Comparison of Artificial Intelligence and Traditional Methods in Preoperative Planning for Primary Total Hip Arthroplasty: A Systematic Review and Meta-Analysis.

Comparison of Artificial Intelligence and Traditional Methods in Preoperative Planning for Primary Total Hip Arthroplasty: A Systematic Review and Meta-Analysis. | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Systematic Review Comparison of Artificial Intelligence and Traditional Methods in Preoperative Planning for Primary Total Hip Arthroplasty: A Systematic Review and Meta-Analysis. Di Xue, Kaiyong Wang, Huan He, Liru Wang, Yupei Dai, Guohang Shen, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5773489/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background The application of artificial intelligence (AI) in orthopedics is becoming increasingly widespread, particularly in the diagnosis and treatment of hip-related diseases. Although AI-assisted total hip arthroplasty (THA) techniques have reached a relatively mature stage, their specific role in preoperative planning for THA remains in the research phase. Current studies are generally small in scale, and their findings appear somewhat fragmented, making it difficult to draw definitive conclusions. Against this backdrop, a systematic review and meta-analysis on the application of AI in THA preoperative planning may provide a more comprehensive and rational answer. Questions/purposes Compared to traditional methods, does artificial intelligence (AI) offer more and better advantages in preoperative planning for patients undergoing primary total hip arthroplasty (THA)? Does it possess potential for future development? Methods We conducted a comprehensive and systematic search in the PubMed, Embase, Web of Science, and Cochrane Library databases, covering the period from their inception to October 30, 2024. This study adheres to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and has been registered in PROSPERO [1] . The included studies focused on patients undergoing primary total hip arthroplasty (THA), with the experimental group using artificial intelligence (AI) for preoperative planning and the control group employing traditional planning methods. We excluded the following: papers published on preprint servers, unpublished studies, conference abstracts, and studies registered on ClinicalTrials.gov but not yet published. Ultimately, data were extracted from 15 eligible studies. To assess the methodological quality of the studies, we applied bias risk assessment methods based on the type of study. The revised Cochrane Risk of Bias tool was employed to assess potential bias in randomized controlled trials (RCTs). For non-randomized controlled trials, including retrospective cohort studies, retrospective case-control studies, and prospective cohort studies, we employed the Newcastle-Ottawa Scale (NOS) for bias risk assessment. Due to the high heterogeneity among studies (I² > 50%), a random-effects model was used for the analysis. Results In the 15 studies that met the inclusion criteria, a total of 2572 participants were included. These patients required primary total hip arthroplasty (THA) due to various hip diseases. Among them, 1307 patients in the experimental group used artificial intelligence (AI) for preoperative planning, while 1265 patients in the control group used traditional methods. There were no statistically significant differences in the baseline characteristics of the included patients (such as age, BMI, preoperative leg length discrepancy, and preoperative Harris score) (P≥0.05), which ensures the reliability of the predictive results. According to the data summary and analysis, compared with traditional methods, AI showed superior performance in the following aspects: the odds ratio (OR) for acetabular component matching accuracy was 0.26 (95% CI, 0.20–0.34; P=0.009; I²=58%), and for femoral component matching accuracy, the OR was 0.25 (95% CI, 0.19–0.32; P=0.66; I²=0%). The matching accuracy was defined with a size difference as the acceptable margin of error. The mean difference (MD) for postoperative leg length discrepancy was -0.49 (95% CI, -0.59 to -0.39; P<0.0001; I²=77%), the MD for surgical time was -16.07 (95% CI, -18.00 to -14.14; P<0.00001; I²=96%), the MD for intraoperative blood loss was -45.91 (95% CI, -61.03 to -30.78; P=0.04; I²=61%), and the MD for postoperative Harris score was 0.83 (95% CI, 0.38–1.28; P=0.001; I²=70%). In addition, the OR for acetabular cup prosthesis prediction accuracy was 0.82 (95% CI, 0.51–1.34; P=0.0001; I²=89%), and the overall average prediction accuracy had an OR of 0.25 (95% CI, 0.18–0.35; P=0.93; I²=0%). Conclusion The results of this systematic review and meta-analysis indicate that artificial intelligence (AI) performs comparably to, or even better than, traditional methods in preoperative planning for hip arthroplasty. Compared with traditional methods, the AI group demonstrated advantages such as reducing surgical time, minimizing intraoperative blood loss, lowering surgical risks, and decreasing surgical trauma. These benefits help promote rapid postoperative recovery, shorten hospital stays, and reduce the occurrence of complications. Additionally, patients in the AI group had higher postoperative Harris scores, less postoperative pain, faster functional recovery, and better postoperative adaptation. AI-assisted preoperative planning for total hip arthroplasty (THA) also improves the accuracy of hip component matching prediction, reduces the likelihood of errors in clinical decision-making, effectively alleviates tensions in the doctor-patient relationship, and reduces the waste of medical resources. Surgery Orthopedics Artificial intelligence (AI) Traditional methods Total hip arthroplasty/Total hip replacement Preoperative planning First-time. Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction Total Hip Arthroplasty (THA) is a surgical procedure primarily used for elderly patients. With the increasing global average life expectancy and the growing proportion of elderly populations, the demand for THA continues to rise. In the United States, from 2020 to 2030, the number of primary and revision THA surgeries is expected to increase by approximately 50% [ 2 , 3 ] . In the United Kingdom, this number is projected to double by 2035 [ 4 ] . Other Western countries, such as Canada, Denmark, and Australia, are also expected to see varying degrees of growth [ 5 – 7 ] . THA has existed in some form for over a century [ 8 , 9 ] , and in recent years, with technological advances and optimized techniques, THA has been dubbed the "surgery of the 21st century," with its unique advantages being evident. THA has achieved significant success in alleviating joint pain, improving function, and enhancing quality of life [ 10 ] , with clinical outcomes being satisfactory, as the prosthetic survival rate over 15–20 years can reach up to 90% [ 11 , 12 ] . As a result, THA has become the preferred treatment for end-stage degenerative joint diseases and hip trauma. By restoring normal anatomical relationships and the biomechanical stability of the hip joint, THA plays a crucial role in improving patients' quality of life [ 13 ] . However, THA is not without its challenges. Prosthetic joint infections, fractures, aseptic loosening, and dislocation [ 14 , 15 ] are the main causes of surgical failure [ 16 ] . These issues may not be entirely avoidable, even for experienced surgeons, and are particularly key obstacles in surgical planning for most general orthopedic surgeons. Therefore, accurate preoperative planning is crucial to improving surgical success rates and reducing complications. Traditional preoperative planning mainly relies on two-dimensional templates [ 17 ] , which are limited by factors such as imaging angles and patient posture variations. As a result, the predicted outcomes are often influenced by the experience level of the surgeon. This reliance on experience often leads to considerable uncertainty, potentially resulting in surgical failure or poor short-term outcomes, which not only causes additional suffering for patients but also wastes valuable medical resources [ 18 ] . Hence, there is an urgent need for an efficient and accurate preoperative planning method. The rapid development of Artificial Intelligence (AI) has introduced new possibilities for preoperative planning in THA. AI's application in the medical field is continuously expanding, with breakthroughs in data analysis and image processing [ 19 ] . In the field of orthopedics, AI's application range is expanding, particularly in the research and treatment of hip diseases. Although current research on AI in THA preoperative planning is still in its early stages, with many studies being small-scale and lacking persuasive power, the potential of AI technology is becoming apparent [ 20 ] . To better understand AI's role in THA preoperative planning, we performed a systematic review and meta-analysis of recent pertinent studies, focusing on the differences between AI and traditional methods in terms of preoperative planning accuracy and postoperative outcomes. By analyzing key indicators such as intraoperative blood loss, surgical time, postoperative leg length discrepancy (LLD), Harris scores, and prosthesis matching accuracy, we aim to address the following questions: Can AI significantly improve the accuracy of preoperative planning? Does it contribute to long-term patient recovery and a reduction in complications? What is its potential for future development? Our analysis may provide strong evidence to support clinical practice. Materials and Methods Our research follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. This study has been registered with the International Prospective Register of Systematic Reviews (PROSPERO) under registration number: CRD42024619714. For the experimental group in our study: The primary method of preoperative planning for total hip arthroplasty (THA) using artificial intelligence (AI) is an intelligent planning system based on 3D imaging and deep learning algorithms. This approach leverages AI technologies, such as deep learning algorithms and image processing techniques, to analyze patient imaging data (e.g., CT, MRI, or X-rays), enabling automated prosthesis selection, position planning, and surgical design. The preoperative planning models employed include AIHIP (Artificial Intelligence Preoperative Planning System for Total Hip Arthroplasty), 3D-AIHIP, robot-assisted THA, and CAD simulative (Computer-Aided Design Simulation). For the control group: The traditional method of preoperative planning for THA is primarily based on the template technique using two-dimensional X-ray images. This method involves obtaining standard X-rays of the patient's hip joint (e.g., anteroposterior pelvic view and lateral hip view), manually analyzing the anatomical features of the hip joint, and using a transparent prosthesis template superimposed on the X-ray to determine the appropriate prosthesis size and model. It also involves planning the implantation position, orientation, and angle of the prosthesis, as well as assessing limb length discrepancy to formulate a basic surgical plan. Data Sources and Searches According to the established search criteria, we conducted a comprehensive search in databases such as Embase, Web of Science, PubMed, and the Cochrane Library. The search date ranged from the establishment of the database to October 30, 2024. We carefully verified the search methods used in each database to avoid any omissions. We also conducted a thorough summary of the search criteria and carefully reviewed the reference lists of relevant studies and reviews to find additional references. In these databases, we identified a total of 388 articles. After excluding 335 articles based on titles and abstracts, we further excluded 16 reviews, 19 duplicates, and 3 articles with inaccessible full texts or data that did not meet the research criteria, leaving 15 articles that were included in the qualitative analysis. The final selection included 2572 participants, with 1307 assigned to the experimental group and 1265 to the control group. The detailed search process can be found in the flowchart (Fig. 1 ). Inclusion and exclusion criteria The method of using the Patient Intervention Compared Outcome of Interest (PICOS) model is as follows: In patients undergoing their first hip replacement surgery (P), artificial intelligence (AI) (I) is compared with traditional methods (C) to assess whether AI is more accurate in preoperative planning than traditional methods (O). The inclusion criteria for our study are: (1) The subjects are patients who require their first hip replacement surgery; (2) The experimental group uses AI models for preoperative planning; (3) The control group uses traditional methods for THA preoperative planning; (4) There are clear intraoperative and/or postoperative recorded indicators; (5) There are no statistically significant differences in baseline information such as age, BMI, preoperative Harris score, and preoperative leg length discrepancy across the studies. Exclusion criteria: (1) Conference abstracts, reviews, and articles that are registered but unpublished; (2) Articles for which the full text is inaccessible; (3) Articles with data that do not meet the research criteria; (4) Articles that are repetitively published (for repetitive publications, we selected the one with the most complete experimental data for statistical analysis). After completing the search, all retrieved articles were first independently reviewed by one researcher (KW) for the titles and abstracts. To prevent selection bias due to factors such as personal fatigue or subjectivity, a second researcher (YD) verified the selection. Any disagreements encountered during this process were discussed, and if unresolved, a third mediator (GS) was involved to reach a consensus. Articles that were deemed appropriate for exclusion were eliminated. Then, the remaining articles were read in full, and those that met the inclusion criteria were finally included in our study. Data Extraction In the final included studies for statistical analysis, data extraction was conducted using Microsoft Excel to create data extraction tables. These tables were finalized after collaborative discussions among multiple team members. In addition to basic information such as title, year, primary author, age, BMI, sample size, and gender ratio, the extracted data primarily focused on intraoperative and postoperative details comparing the artificial intelligence group and the traditional method group. Key data points included the accuracy of acetabular and femoral matching predictions, operative time, intraoperative blood loss, postoperative leg length discrepancy, postoperative Harris scores, acetabular cup prediction accuracy, preoperative and postoperative VAS scores, postoperative inclination angle comparison, and length of hospital stay. Two researchers (KW and YD) independently conducted the data extraction process. Any discrepancies that occurred during the extraction process were addressed through discussion and a review by a third reviewer (YC). If disagreements persisted, more experienced researchers (DX) were consulted for guidance until a consensus was reached. (Table 1 .) Table 1 Basic information on preoperative planning for THA between the artificial intelligence group and the traditional methods group. Source (country) Sex (male/ female) Age: Mean ± SD (year) Follow up(m) BMI(kg/㎡) Inclusion model Postoperative Harris hip score Number of participants AI control AI control AI control AI control AI control Zheng, H.2024 (China) NR 64.3 ± 10 61.3 ± 10.2 NR 24.4 ± 2.4 23.8 ± 2.5 AIHIP 2D model 95.3 ± 3.1 96.0 ± 2.0 30 30 Yang, W.2024 (China) 193/247 56.6 ± 14.5 57 ± 15.2 NR NR NR AIHIP 2D model 85.51 ± 3.69 84.88 ± 2.68 220 220 Xie, H.2024 (China) 16/87 42.3 ± 12.4 42.3 ± 12.4 NR 24.0 ± 2.7 24.0 ± 2.7 AIHIP 2D model NR NR 164 164 Sun, G. Y.2024 (China) 32/40 62.2 ± 10.9 60.9 ± 12.1 14.5 ± 2.1 NR NR AIHIP 2D model 92.1 ± 5.8 90.5 ± 6.2 36 36 Meipeng, M.2024 (China) 28/32 70.2 ± 4.6 69.8 ± 5.4 5.5(4–6) 22.8 ± 4.2 23.1 ± 3.7 AIHIP 2D model 81.43 ± 4.65 77.51 ± 3.28 30 30 Kai, Z.2024 (China) 73/86 59.83 ± 11.89 61.90 ± 13.88 13.72 ± 5.65 22.20 ± 1.04 22.50 ± 1.75 AIHIP 2D model 92.74 ± 3.08 91.81 ± 3.52 80 79 Cheng, K.2024 (China) 30/32 60.5 ± 13.2 60.2 ± 12.5 NR NR NR CAD simulative routine X-ray 81.9 ± 6.5 74.7 ± 11.1 31 31 Anwar, A.2024 (China) 52/65 62.3 ± 10.7 62.3 ± 10.7 NR 25.5 ± 3.4 25.5 ± 3.4 AIHIP 2D model NR NR 117 117 Zhang, S.2023 (China) 19/48 46 ± 18.5 45 ± 17.5 12 25.3 ± 8.9 24.6 ± 7.6 robot-assisted THA manual THA 83 ± 13 83 ± 11.5 36 31 Wu, L.2023 (China) 101/60 57.6 ± 10.5 57.6 ± 10.5 NR 25.3 ± 3.0 24.7 ± 3.8 3D-AIHIP 2D model NR NR 95 66 Wu, L.2023 (China) 31/30 58.2 ± 9.8 60.5 ± 11.1 >12 25.9 ± 2.6 26.1 ± 3.8 3D-AIHIP 2D model 86.38 ± 4.04 84.37 ± 5.30 34 27 Mieradili, Maimaitiyiming 2023(China) 10/18. 59.3 ± 16.8 59.3 ± 16.8 NR NR NR AIHIP X-ray model NR NR 28 28 Chen, X.2022 (China) 61/59 47.62 ± 15.30 53.75 ± 16.10 3 24.19 ± 3.08 25.14 ± 3.78 AIHIP X-ray model NR NR 60 60 Ding, X. Z.2021 (China) 192/124 50.68 ± 12.64 50.68 ± 12.64 NR 25.07 ± 3.20 25.07 ± 3.20 3D-AIHIP 2D model NR NR 316 316 Wu, D.2020 (China) 31/29 46.9 ± 12.6 47.0 ± 21.8 NR NR NR 3D-AIHIP film template NR NR 30 30 SD: standard deviation; BMI: body mass index; NR: no reports; AIHIP: artificial intelligence preoperative planning system for total hip arthroplasty; CAD simulative: computer-aided design simulation; robot-assisted THA: robot-assisted total hip arthroplasty; 3D-AIHIP: artificial intelligence-assisted three-dimensional preoperative planning system for total hip arthroplasty. Assessment of Article Quality and Risk of Bias In our study data, four studies were randomized controlled trials (RCTs), while the remaining studies were non-randomized controlled trials, including six retrospective cohort studies, two case-control studies, and three prospective cohort studies. Therefore, we adopted two different evaluation methods for assessing the quality of the articles. For the RCTs, the revised Cochrane risk-of-bias tool for randomized trials (ROB 2) was used, while the Newcastle-Ottawa Scale (NOS) was employed to evaluate the risk of bias for the non-randomized intervention studies included in our analysis. Statistical analysis To compare artificial intelligence (AI) with traditional methods in THA preoperative planning, we summarized the data for the experimental and control groups in each study and conducted a statistical analysis using a random-effects model. Since the collected data included both binary variables and continuous variables, in our analysis, we created forest plots that displayed either odds ratios (OR) for binary variables or mean differences (MD) for continuous variables. Heterogeneity was evaluated using the I² statistic, where I² > 50% was considered indicative of substantial heterogeneity, and I² ≤ 50% indicated low heterogeneity. We illustrated the risk of bias using risk-of-bias plots and bar charts. The odds ratios (OR) or mean differences (MD) between the AI and traditional method groups were calculated with 95% confidence intervals (95% CI). The data were analyzed using RevMan version 5.4. Results We included 15 studies [ 21 – 35 ] . Among the four randomized controlled trials (RCTs), two had no high-risk biases, while the other two each had one high-risk bias in terms of incomplete outcome data and selective reporting bias. Overall, the quality of the RCTs was quite high. For the 11 non-randomized controlled trials included, the NOS scores ranged from 4 to 8, with most scoring between 6 and 7. The NOS scale ranges from 0 to 9, and studies scoring above 4 are considered of moderate to high quality. Thus, the quality of the included studies was relatively high. Details can be found in the risk-of-bias assessment chart (Fig. 2 ). Since the baseline characteristics during preoperative planning showed no statistical differences across studies, the postoperative statistical data are more comparable. We conducted multiple subgroup analyses of the intraoperative and postoperative data included in the studies. The results generally indicated that the AI group demonstrated advantages over the traditional method group in terms of preoperative planning accuracy, reducing surgery time, and lowering surgical risks. Prediction Accuracy for Acetabular and Femoral Sides: The comparison is made between the preoperative planned prosthesis and the prosthesis actually used during the surgery. If the two are an exact match or differ by no more than one size (in THA, a difference of one size between the predicted acetabular or femoral prosthesis and the actual prosthesis used during the surgery is generally considered acceptable; the size difference of one size varies depending on the prosthesis type and the design specifications of different manufacturers, typically ranging from 1 to 2 millimeters), they are considered to have an accurate match. The predictive matching accuracy for the acetabular side [22,23,25–29,31−34] in the AI group compared to the traditional method showed an OR of 0.26 (95% CI, 0.20–0.34). For the femoral side prediction accuracy [22,23,25–29,31−34] , the OR was 0.25 (95% CI, 0.19–0.32). In both cases, the odds ratio (OR) for the experimental group compared to the control group was less than 1, indicating fewer prediction errors in the AI group, which suggests superior accuracy. (Fig. 3 .) Postoperative Leg Length Discrepancy (LLD): LLD refers to the length discrepancy between the lower limbs on both sides of the patient after THA. Specifically, it refers to the difference in skeletal length between the operated leg and the opposite leg, which is typically assessed through imaging examinations (such as X-rays or CT scans) or physical measurements (such as bedside measurement). A postoperative length difference of 1–2 centimeters is generally considered acceptable. Nine studies [ 21 , 23 , 25 – 28 , 30 , 31 , 33 ] reported postoperative leg length discrepancy (LLD). The mean difference (MD) between the experimental group and the control group was − 0.49 (95% CI, -0.59 to -0.39). This indicates that preoperative planning with artificial intelligence resulted in lower leg length discrepancies postoperatively compared to the control group. The surgical error margin in the experimental group was smaller, leading to better postoperative adaptation for patients. (Fig. 4 .) Surgical Time and Intraoperative Blood Loss: Eight studies included data on surgical time [ 21 , 23 , 25 , 26 , 29 – 31 , 33 ] , and five studies [ 21 , 23 , 25 , 29 , 33 ] reported intraoperative blood loss. The mean differences (MD) between the experimental group and the control group were − 16.07 (95% CI, -18.00 to -14.00) and − 45.91 (95% CI, -61.03 to -30.78), respectively. These results indicate that preoperative planning with artificial intelligence (AI) for THA, compared to traditional methods, resulted in shorter surgical times and reduced intraoperative blood loss. Shorter surgical times can effectively lower surgical risks, including reducing the likelihood of infections caused by prolonged exposure. Decreased intraoperative blood loss reduces patient life risk and avoids unnecessary blood transfusions, thereby mitigating transfusion-related risks such as antibody reactions. Furthermore, this reduction helps alleviate blood supply shortages, ensuring that more patients in critical need of transfusions can access the necessary resources. (Fig. 5 .) Postoperative Harris Score (HHS): HHS is a standardized scoring system used to assess the recovery of hip joint function in patients after total hip arthroplasty (THA). This scoring system takes into account the patient's level of pain, activity ability, gait, joint function and range, as well as the ability to perform daily activities related to the hip joint. The total score ranges from 0 to 100 and is interpreted as follows: 90–100 points: indicates good function with almost no issues. 80–89 points: indicates fair function, with possible mild symptoms. 70–79 points: indicates moderate function, with some discomfort and limitations. Below 60 points: indicates poor function, with significant pain or functional impairment. In other words, a higher score indicates better postoperative recovery. In this research, eight studies [ 21 , 23 , 25 , 27 – 31 ] included the postoperative Harris scores (and the preoperative Harris scores of each study showed no statistical difference, indicating that the postoperative Harris scores are more comparable). The mean difference (MD) of postoperative Harris scores in these 8 studies was 0.83 (95% CI, 0.38–1.28), which means that the average postoperative Harris score in the AI group was higher than in the traditional method group. The I² was 70%, indicating significant heterogeneity, with a P-value of < 0.05, which shows a statistically significant difference. This suggests that preoperative planning using AI results in less pain and better functional recovery postoperatively, leading to a more comfortable treatment experience for the patients. (Fig. 6 .) Accuracy of Acetabular Cup Prosthesis Prediction: The accuracy of acetabular cup prosthesis prediction refers to the degree of agreement between the size, position, and angle of the acetabular cup prosthesis predicted in the preoperative plan and the actual prosthesis implanted during the surgery in THA. If the size, position, and angle deviations of the acetabular cup prosthesis are within an acceptable range (which may vary depending on the design specifications of different manufacturers, with a general range being: size deviation of 1-2mm, position deviation of 5mm, and angle deviation of 5°), the preoperative planned prosthesis is considered to match the actual implanted prosthesis accurately. The higher the accuracy, the smaller the deviation between the preoperative plan and the actual intraoperative operation, resulting in better surgical outcomes. There are three studies [ 21 , 25 , 28 ] evaluated the accuracy of acetabular cup prosthesis prediction, with an odds ratio (OR) of 0.82 (95% CI, 0.51–1.34) and P = 0.43. Although the small sample size limits the persuasiveness of the findings, the overall trend suggests that the AI group demonstrates superior accuracy in predicting acetabular prosthesis outcomes compared to the traditional method group. (Fig. 7 .) Overall, the AI group has indeed brought significant surprises in the field of healthcare. However, AI still cannot guarantee complete accuracy, but its potential for development is vast. Discussion Advantages and challenges of preoperative planning assisted by AI in THA: Total Hip Arthroplasty (THA) is currently the primary surgical method for treating persistent pain and functional limitations caused by advanced hip joint diseases. With the increasing global average lifespan and continuous advancements in surgical technology, the age range of patients undergoing THA has expanded significantly, showing a growing trend of polarization [ 2 , 22 ] . This shift not only imposes higher demands on surgical complexity but also profoundly impacts traditional medical decision-making models [ 9 ] . It is projected that the number of patients requiring THA annually will increase substantially in the future. Although THA is one of the most successful procedures in orthopedics, its outcomes can still be significantly affected by the choice of prosthesis model and the precision of implantation. Improper prosthesis matching can lead to intraoperative and postoperative complications such as early prosthesis loosening, dislocation, or even surgical failure, ultimately affecting the longevity of the prosthesis and the patient's mobility [ 36 ] . Precise matching between the prosthesis and bone, optimal implant positioning, and superior friction interfaces are key to constructing a stable mechanical structure and prolonging the lifespan of the prosthesis. To achieve these goals, comprehensive and accurate preoperative planning is crucial. Preoperative planning optimizes surgical processes, reduces unnecessary steps during surgery, shortens operation time, lowers risks, and promotes postoperative recovery [ 21 , 24 ] . In recent years, with the rise of digital orthopedic technology, three-dimensional preoperative planning based on CT or X-ray imaging data has demonstrated significant advantages in the treatment of hip joint diseases. Research has demonstrated that AI-assisted 3D preoperative design techniques can facilitate more accurate, safer, and consistently reproducible acetabular prosthesis implantation [ 26 ] . Limitations of traditional preoperative planning and the rise of artificial intelligence: Traditional manual imaging review heavily relies on the personal experience of physicians, leading to significant variations in preoperative planning accuracy. In contrast, Artificial Intelligence (AI) with deep learning technology can integrate diagnostic expertise from multiple specialists, offering more objective and precise diagnostic and planning recommendations through big data analysis. Studies have shown that, compared to traditional methods, AI excels in the predictive accuracy of prosthesis models and implant angles. For instance, AI-assisted three-dimensional preoperative planning achieves prediction accuracies of 90% for acetabular cups and 83% for femoral stems, significantly higher than the 57% and 53% achieved by traditional two-dimensional X-ray planning [ 25 ] .In our study, the AI-assisted group not only significantly improved prosthesis matching accuracy but also outperformed traditional methods in several key metrics, including shorter surgery time, reduced intraoperative blood loss, smaller postoperative leg length discrepancy (LLD), and higher postoperative Harris scores. Particularly in prosthesis implant angle planning, AI, combined with 3D printing technology and personalized navigation templates, can reconstruct individualized surgical models, optimizing implant paths and angles to minimize fluoroscopy use and surgical trauma [ 23 ] . The potential of AI in managing THA postoperative complications: The hip joint bears the majority of daily weight-bearing activities, and improper prosthesis matching can lead to postoperative complications such as early dislocation, pain, and loosening [ 37 ] . AI-assisted preoperative planning enables more accurate predictions of prosthesis models and implant angles, achieving optimal matching between the prosthesis and bone, thereby reducing the incidence of dislocations and other complications, while enhancing joint stability and prosthesis longevity [ 24 ] . LLD is one of the most common postoperative complications, considered a primary cause of postoperative pain, gait instability, and aseptic loosening. In severe cases, it may necessitate revision surgery within a short period [ 23 , 38 ] . Statistical analysis indicates that patients with AI-assisted planning exhibit significantly lower postoperative LLD means compared to the traditional method group (MD = -0.49, 95% CI: -0.59 to -0.39, P < 0.05), demonstrating statistically significant differences. Furthermore, regarding the precision of acetabular prosthesis implantation, the proportion of AI-planned acetabular prostheses positioned within Lewinnek and Callanan safe zones reached 86.32% and 83.2%, respectively, significantly exceeding the 72.73% and 69.7% achieved by traditional methods. This highlights AI's advantages in planning complex cases. Challenges and future perspectives: Although AI demonstrates numerous advantages in preoperative planning for Total Hip Arthroplasty (THA), certain challenges remain in its clinical application. First, current studies are limited in sample size and follow-up duration, and the lack of long-term follow-up data makes it difficult to comprehensively evaluate AI's impact on prosthesis longevity and functional recovery. Second, AI's development in the medical field is still in its early stages, and its reliability and generalizability in complex cases require further validation. In conclusion, AI offers more precise and efficient solutions for THA preoperative planning, with remarkable potential to optimize surgical workflows, enhance postoperative functional recovery, and reduce complications. However, the widespread adoption of AI will require large-scale, long-term follow-up studies and multidisciplinary collaboration to continually refine the technology and optimize its application strategies, paving the way for AI to achieve greater maturity in the field of orthopedics. Limitations Our study has several limitations that warrant attention and improvement in future research. First, the follow-up duration in the included studies was limited, with the longest recorded follow-up being 19.37 months [ 28 ] and the shortest only 3 months. Although the AI-assisted preoperative planning group demonstrated superior performance in prosthesis placement accuracy and correction of postoperative leg length discrepancy (LLD), it remains unclear whether these advantages in prosthesis accuracy can be sustained over medium- to long-term follow-ups. Therefore, future research should involve larger sample sizes and longer follow-up periods to validate the long-term benefits of these advantages. Second, all the studies we included were conducted in China. Although this reflects the objective findings of our screening process, the single-region research background may pose certain limitations. To ensure the broader applicability of our findings, we adhered to strict bias risk assessment methods, using the Cochrane risk of bias tool for randomized controlled trials (RCTs) and the Newcastle-Ottawa Scale (NOS) for non-randomized controlled trials (non-RCTs). Fortunately, the overall quality of the included studies was relatively high. However, research conducted in a single region may not fully represent the situation in different parts of the world. Therefore, we look forward to the inclusion of more relevant studies from diverse regions in the future to enrich the research content and enhance the reliability and generalizability of the conclusions. Conclusion Compared to traditional preoperative planning methods, artificial intelligence (AI)-assisted preoperative planning demonstrates significant advantages. Firstly, AI provides a more intuitive visualization of the anatomical structure of the affected area, combined with three-dimensional technology for preoperative simulations. This enhances surgical precision and reduces procedural complexity. Additionally, the application of AI decreases the need for intraoperative fluoroscopy and repeated prosthesis measurements. These benefits not only shorten the operation time and minimize intraoperative blood loss but also effectively lower surgical risks and trauma, promoting faster postoperative recovery, reducing the incidence of complications, and ultimately preventing the wastage of medical resources. In terms of prosthesis parameter prediction and matching accuracy, AI surpasses traditional methods. This benefit aids in ensuring both the initial stability and the long-term durability of the prosthesis [ 22 ] . However, to further confirm the reliability and feasibility of these conclusions, future studies need larger-scale randomized controlled trials (RCTs) and extended follow-up periods to provide stronger evidence.AI has demonstrated extensive potential in the medical field, with vast opportunities for further development in both depth and breadth. However, it is crucial to recognize that technology is a double-edged sword. No matter how advanced AI becomes, its development and application should always be guided by principles of humanistic ethics, ensuring that it serves the health and well-being of humanity and society. Declarations Supplementary Information Supplementary materials mentioned in the article can be found in the appendix. Authors’ Contributions Kaiyong Wang, Yupei Dai and Di Xue wrote the first draft of the article, Kaiyong Wang,Yupei Dai and Guohang Shen made revisions to the manuscript, as well as language polishing of the article, Guohang Shen, Yang Chen assisted in the process of data extraction and entry, Kaiyong Wang,Yupei Dai and Yang Chen provided reference materials, and Kaiyong Wang and Di Xue finally read and confirmed the manuscript of the article. Funding Natural Science Foundation of Ningxia Hui Autonomous Region（2024AAC03601，2024AAC03663，2024AAC03665），Ningxia Medical University project（XT2023035），The central government of Ningxia Hui Autonomous Region guides the special project of local science and technology development（2024FRD05048，2024FRD05108），Key R&D project of Ningxia Hui Autonomous Region（2022BEG03126，2022BEG03169）. Ethical Statement: This study falls within the scope of retrospective research and does not involve direct human participants, animal experiments, or the collection and use of sensitive data. Therefore, no ethical issues are present. Acknowledgements None. Conflict of Interest The authors declare that there are no conflicts of interest associated with the research presented in this paper. References Lan Z, Lin X, Xue D et al (2024) Can Bisphosphonate Therapy Reduce Overall Mortality in Patients With Osteoporosis? A Meta-analysis of Randomized Controlled Trials. Clinical orthopaedics and related research http://dx.doi.org/10.1097/corr.0000000000003204 Kurtz S, Ong K, Lau E et al (2007) Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am Vol 89(4):780–785. http://dx.doi.org/10.2106/jbjs.F.00222 Sloan M, Premkumar A, Sheth NP (2018) Projected Volume of Primary Total Joint Arthroplasty in the U.S., 2014 to 2030. J Bone Joint Surg Am Vol 100:17, 1455–1460. http://dx.doi.org/10.2106/jbjs.17.01617 Culliford D, Maskell J, Judge A et al (2015) Future projections of total hip and knee arthroplasty in the UK: results from the UK Clinical Practice Research Datalink. Osteoarthr Cartil 23:4, 594–600. http://dx.doi.org/10.1016/j.joca.2014.12.022 Sharif B, Kopec J, Bansback N et al (2015) Projecting the direct cost burden of osteoarthritis in Canada using a microsimulation model. Osteoarthr Cartil 23:10, 1654–1663. http://dx.doi.org/10.1016/j.joca.2015.05.029 Nemes S, Gordon M, Rogmark C et al (2014) Projections of total hip replacement in Sweden from 2013 to 2030. Acta Orthop 85:3, 238–243. http://dx.doi.org/10.3109/17453674.2014.913224 Ackerman IN, Bohensky MA, Zomer E et al (2019) The projected burden of primary total knee and hip replacement for osteoarthritis in Australia to the year 2030. BMC Musculoskelet Disord 20(1):90. http://dx.doi.org/10.1186/s12891-019-2411-9 Knight SR, Aujla R, Biswas SP (2011) Total Hip Arthroplasty - over 100 years of operative history. Orthop Rev 3:e16. 2 http://dx.doi.org/10.4081/or.2011.e16 Learmonth ID, Young C, Rorabeck C (2007) The operation of the century: total hip replacement. Lancet (London England) 370:9597, 1508–1519. http://dx.doi.org/10.1016/s0140-6736(07)60457-7 Imamura M, Munro NA, Zhu S et al (2012) Single mini-incision total hip replacement for the management of arthritic disease of the hip: a systematic review and meta-analysis of randomized controlled trials. J Bone Joint Surg Am Vol 94:20, 1897–1905. http://dx.doi.org/10.2106/jbjs.K.00495 Alp NB, Akdağ G, Erdoğan F (2020) Long-term results of total hip arthroplasty in developmental dysplasia of hip patients. Joint Dis Relat Surg 31:2, 298–305. http://dx.doi.org/10.5606/ehc.2020.74412 Evans JT, Evans JP, Walker RW et al (2019) How long does a hip replacement last? A systematic review and meta-analysis of case series and national registry reports with more than 15 years of follow-up. Lancet (London England) 393:10172, 647–654. http://dx.doi.org/10.1016/s0140-6736(18)31665-9 Kumar Vyas A, Gupta A, Patni P et al (2020) Evaluation of results of various tendon transfers in high and low radial nerve palsy. J Clin Orthop trauma 11:4, 614–619. http://dx.doi.org/10.1016/j.jcot.2020.05.023 Karachalios T, Komnos G, Koutalos A (2018) Total hip arthroplasty: Survival and modes of failure. EFORT open reviews 3:5, 232–239. http://dx.doi.org/10.1302/2058-5241.3.170068 Di Martino A, Brunello M, Bordini B et al (2023) Unstable Total Hip Arthroplasty: Should It Be Revised Using Dual Mobility Implants? A Retrospective Analysis from the R.I.P.O. Registry. J Clin Med 12:2. http://dx.doi.org/10.3390/jcm12020440 Viceconti M, Ascani D, Mazzà C (2019) Pre-operative prediction of soft tissue balancing in knee arthoplasty part 1: Effect of surgical parameters during level walking. J Orthop research: official publication Orthop Res Soc 37:7, 1537–1545. http://dx.doi.org/10.1002/jor.24289 Ho CWG, Tan BK, Ong BH et al (2021) Customized Two-Dimensional Computed Tomography-Guided Preoperative Pectus Bar Shaping. Annals of plastic surgery 87. 6676–680. http://dx.doi.org/10.1097/sap.0000000000002892 Gu LC, Peng Y, Zhang Y et al (2024) Enhancing treatment outcomes for Acute Periprosthetic Hip Joint infection: optimizing debridement, antibiotics, and Implant Retention through vacuum sealing drainage in the deep tissue. Arch Orthop Trauma Surg 145:1, 54. http://dx.doi.org/10.1007/s00402-024-05649-z Shen D, Wu G, Suk HI (2017) Deep Learning in Medical Image Analysis. Annu Rev Biomed Eng 19:221–248. http://dx.doi.org/10.1146/annurev-bioeng-071516-044442 Zaharchuk G (2019) Next generation research applications for hybrid PET/MR and PET/CT imaging using deep learning. Eur J Nucl Med Mol Imaging 46:13, 2700–2707. http://dx.doi.org/10.1007/s00259-019-04374-9 Zheng H, Feng E, Xiao Y et al (2024) Is AI 3D-printed PSI an accurate option for patients with developmental dysplasia of the hip undergoing THA? BMC Musculoskelet Disord 25(1):308. http://dx.doi.org/10.1186/s12891-024-07449-3 Anwar A, Zhang Y, Zhang Z et al (2024) Artificial intelligence technology improves the accuracy of preoperative planning in primary total hip arthroplasty. Asian J Surg 47:7, 2999–3006. http://dx.doi.org/10.1016/j.asjsur.2024.01.133 Yang W, Gao T, Liu X et al (2024) Clinical application of artificial intelligence-assisted three-dimensional planning in direct anterior approach hip arthroplasty. Int Orthop 48:3, 773–783. http://dx.doi.org/10.1007/s00264-023-06029-9 Xie H, Yi J, Huang Y et al (2024) Application and evaluation of artificial intelligence 3D preoperative planning software in developmental dysplasia of the hip. J Orthop Surg Res 19:1, 176. http://dx.doi.org/10.1186/s13018-024-04588-0 Meipeng M, Jin W, Rafi U et al (2024) Role and significance of artificial intelligence preoperative planning in total hip arthroplasty. Chin J Tissue Eng Res 28:9, 1372–1377. http://dx.doi.org/10.12307/2023.798 Wu L, Zhao X, Lu ZD et al (2023) Accuracy analysis of artificial intelligence-assisted three-dimensional preoperative planning in total hip replacement. Joint Dis Relat Surg 34:3, 537–547. http://dx.doi.org/10.52312/jdrs.2023.1059 Sun GY, Jiang YK, Li T et al (2024) AI-HIP system for prosthesis size, global femoral offset and osteotomy in total hip arthroplasty. Zhongguo gu shang = China J Orthop Traumatol 37:9, 848–854. http://dx.doi.org/10.12200/j.issn.1003-0034.20240276 Kai Z, Zhuotao G, Qiaoqiao M et al (2024) Accuracy and influencing factor of artificial intelligence planning system in patients undergoing total hip arthroplasty. Chin J Tissue Eng Res 28:12, 1863–1868. http://dx.doi.org/10.12307/2024.026 Cheng K, Zhu HT, Peng YH et al (2024) To further incorporate computer-aided designs to improve preoperative planning in total hip arthroplasty: a cohort study. Front Surg 11. http://dx.doi.org/10.3389/fsurg.2024.1345261 Zhang S, Liu Y, Ma M et al (2023) Is Robotic-Assisted Technology Still Accurate in Total Hip Arthroplasty for Fibrous-Fused Hips? J Arthroplast 38:1, 129–134. http://dx.doi.org/10.1016/j.arth.2022.07.023 Wu L, Yang XC, Wu J et al (2023) Short-term outcome of artificial intelligence-assisted preoperative three-dimensional planning of total hip arthroplasty for developmental dysplasia of the hip compared to traditional surgery. 34:571–582. Joint diseases and related surgery http://dx.doi.org/10.52312/jdrs.2023.1076 Mieradili M, Yilihamujiang W, Sun R et al (2023) : Application of artificial intelligence preoperative planning system in total hip arthroplasty for adult developmental dysplasia of the hip. Zhongguo xiu fu chong jian wai ke za zhi = Zhongguo xiufu chongjian waike zazhi = Chinese journal of reparative and reconstructive surgery 37, 1, 25–30. http://dx.doi.org/10.7507/1002-1892.202209098 Chen X, Liu X, Wang Y et al (2022) Development and Validation of an Artificial Intelligence Preoperative Planning System for Total Hip Arthroplasty. Front Med 9:841202. http://dx.doi.org/10.3389/fmed.2022.841202 Ding XZ, Zhang BS, Li WA et al (2021) Value of preoperative three-dimensional planning software (AI-HIP) in primary total hip arthroplasty: a retrospective study. J Int Med Res 49:11. http://dx.doi.org/10.1177/03000605211058874 Wu D, Liu X, Zhang Y et al (2020) : Research and application of artificial intelligence based three-dimensional preoperative planning system for total hip arthroplasty. Zhongguo xiu fu chong jian wai ke za zhi = Zhongguo xiufu chongjian waike zazhi = Chinese journal of reparative and reconstructive surgery 34, 9, 1077–1084. http://dx.doi.org/10.7507/1002-1892.202005007 Callanan MC, Jarrett B, Bragdon CR et al (2011) The John Charnley Award: risk factors for cup malpositioning: quality improvement through a joint registry at a tertiary hospital. Clin Orthop Relat Res 469:2, 319–329. http://dx.doi.org/10.1007/s11999-010-1487-1 Wang J, Min L, Lu M et al (2020) : What are the Complications of Three-dimensionally Printed, Custom-made, Integrative Hemipelvic Endoprostheses in Patients with Primary Malignancies Involving the Acetabulum, and What is the Function of These Patients? Clinical orthopaedics and related research 478, 11, 2487–2501. http://dx.doi.org/10.1097/corr.0000000000001297 Ofa SA, Ross AJ, Ross BJ et al (2021) : Complication Rates of Hemiarthroplasty Conversion to Total Hip Arthroplasty Versus Primary Total Hip Arthroplasty. Orthopedic reviews 13, 2, 25539. http://dx.doi.org/10.52965/001c.25539 Additional Declarations The authors declare no competing interests. Supplementary Files Appendix.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5773489","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Systematic Review","associatedPublications":[],"authors":[{"id":398272943,"identity":"96d2a6ef-6594-4115-8043-32633504a378","order_by":0,"name":"Di Xue","email":"","orcid":"","institution":"Ningxia Key Laboratory of Clinical and Pathogenic Microbiology Institute of Medical Sciences, General Hospital of Ningxia Medical University","correspondingAuthor":false,"prefix":"","firstName":"Di","middleName":"","lastName":"Xue","suffix":""},{"id":398272944,"identity":"42ebcd35-ae0f-4e16-bd4d-c0672e6b9184","order_by":1,"name":"Kaiyong Wang","email":"","orcid":"","institution":"Ningxia Key Laboratory of Clinical and Pathogenic Microbiology Institute of Medical Sciences, General Hospital of Ningxia Medical University","correspondingAuthor":false,"prefix":"","firstName":"Kaiyong","middleName":"","lastName":"Wang","suffix":""},{"id":398272947,"identity":"93242f78-f60f-42cb-8984-841b40222eab","order_by":2,"name":"Huan He","email":"","orcid":"","institution":"Ningxia Key Laboratory of Clinical and Pathogenic Microbiology Institute of Medical Sciences, General Hospital of Ningxia Medical University","correspondingAuthor":false,"prefix":"","firstName":"Huan","middleName":"","lastName":"He","suffix":""},{"id":398272948,"identity":"a57baf0c-499c-40b6-8960-b466ffe90d16","order_by":3,"name":"Liru Wang","email":"","orcid":"","institution":"Department of Clinical Laboratory, General Hospital of Ningxia Medical University","correspondingAuthor":false,"prefix":"","firstName":"Liru","middleName":"","lastName":"Wang","suffix":""},{"id":398272949,"identity":"53dda546-1f88-46f9-aa23-30d0a0f74b92","order_by":4,"name":"Yupei Dai","email":"","orcid":"","institution":"The third ward of orthopaedic department, General Hospital of Ningxia Medical University, 804 Shengli South Street, Yinchuan, Ningxia Hui Autonomous Region 750004, PR China.","correspondingAuthor":false,"prefix":"","firstName":"Yupei","middleName":"","lastName":"Dai","suffix":""},{"id":398272950,"identity":"1c393e02-f4f3-4a0e-b1af-8909da299230","order_by":5,"name":"Guohang Shen","email":"","orcid":"","institution":"The third ward of orthopaedic department, General Hospital of Ningxia Medical University, 804 Shengli South Street, Yinchuan, Ningxia Hui Autonomous Region 750004, PR China.","correspondingAuthor":false,"prefix":"","firstName":"Guohang","middleName":"","lastName":"Shen","suffix":""},{"id":398272951,"identity":"b41d549e-b7af-4dfc-aa20-08b91f0ed076","order_by":6,"name":"Yang Chen","email":"","orcid":"","institution":"The third ward of orthopaedic department, General Hospital of Ningxia Medical University, 804 Shengli South Street, Yinchuan, Ningxia Hui Autonomous Region 750004, PR China.","correspondingAuthor":false,"prefix":"","firstName":"Yang","middleName":"","lastName":"Chen","suffix":""},{"id":398272952,"identity":"1d1843cd-25cf-42e1-9f41-d67b6cd7bfa6","order_by":7,"name":"Yajing Su","email":"","orcid":"","institution":"Ningxia Key Laboratory of Clinical and Pathogenic Microbiology Institute of Medical Sciences, General Hospital of Ningxia Medical University","correspondingAuthor":false,"prefix":"","firstName":"Yajing","middleName":"","lastName":"Su","suffix":""},{"id":398272953,"identity":"6b5030bb-80c0-40e4-ba85-5f7abc6d6c5d","order_by":8,"name":"Xue Lin","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA30lEQVRIie2QPQrCQBBGJwYWi9F0ErDYK0xIYxG8gleIBKxzhEhaJa1BwSsoucDKgpWewEawtViwSSHoKpZitrTY18xXzGN+ACyW/8QFoGioqxCqNlbSSQLAxrtyZqwo6WSAoWwzg34+OoQhkuvqoCQgcK8nfiuBOIQJEusG4riR6QCCchk3KNN5JZHQ1WEjFwgxnZqUvPNSfGeb41kiM1A461R6MXLWDMFMIfRuwYrihJCRfrLffAsv9ol/vT+GvJAXpeqIe/2mKQKghZ+g8X+3v6dkAE79CRaLxWL5yhNb80QRCYWRFgAAAABJRU5ErkJggg==","orcid":"","institution":"Ningxia Key Laboratory of Clinical and Pathogenic Microbiology Institute of Medical Sciences, General Hospital of Ningxia Medical University","correspondingAuthor":true,"prefix":"","firstName":"Xue","middleName":"","lastName":"Lin","suffix":""}],"badges":[],"createdAt":"2025-01-06 11:49:39","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":true,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":true},"doi":"10.21203/rs.3.rs-5773489/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5773489/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":73199870,"identity":"662e093f-35fa-44b6-bc29-bf3831265620","added_by":"auto","created_at":"2025-01-07 16:06:50","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":204870,"visible":true,"origin":"","legend":"\u003cp\u003eSelection process for systematic review.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-5773489/v1/052add43d12e958fabcae53f.png"},{"id":73201011,"identity":"9e21195f-6557-4e3d-a1ac-00f29b944811","added_by":"auto","created_at":"2025-01-07 16:14:50","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":374842,"visible":true,"origin":"","legend":"\u003cp\u003eRisk of bias assessment chart.\u003c/p\u003e\n\u003cp\u003eA and B are bias assessments for Randomized Controlled Trials (RCTs).\u003c/p\u003e\n\u003cp\u003eC is a bias assessment for non-Randomized Controlled Trials (non-RCTs).\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-5773489/v1/d842df3d86c6e69052adb241.png"},{"id":73199875,"identity":"97882921-baa3-4dd4-a297-50f155f8b896","added_by":"auto","created_at":"2025-01-07 16:06:50","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":197254,"visible":true,"origin":"","legend":"\u003cp\u003eA: Forest plot of acetabular side matching accuracy.\u003c/p\u003e\n\u003cp\u003eB: Funnel plot of acetabular side matching accuracy.\u003c/p\u003e\n\u003cp\u003eC: Forest plot of femoral side matching accuracy.\u003c/p\u003e\n\u003cp\u003eD: Funnel plot of femoral side matching accuracy.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-5773489/v1/8aefd7a348fd932444a83141.png"},{"id":73199882,"identity":"8b93b9a7-8aa9-4fd6-a165-8610ef5e4645","added_by":"auto","created_at":"2025-01-07 16:06:50","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":135988,"visible":true,"origin":"","legend":"\u003cp\u003eLeg length discrepancy (LLD).\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-5773489/v1/061f7b2b7ff8fe0b7b30d748.png"},{"id":73199888,"identity":"a6a9d54e-4396-40d6-aeb8-2ae1715723fb","added_by":"auto","created_at":"2025-01-07 16:06:51","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":251032,"visible":true,"origin":"","legend":"\u003cp\u003eA: Time of operation\u003c/p\u003e\n\u003cp\u003eB: Blood loss volume\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-5773489/v1/b88e16d5ba8e9d3698ab204b.png"},{"id":73199893,"identity":"d852d9e3-2883-4d61-beb8-bebc733bf7ca","added_by":"auto","created_at":"2025-01-07 16:06:51","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":129389,"visible":true,"origin":"","legend":"\u003cp\u003ePostoperative Harris hip score (HHS).\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-5773489/v1/edf48b211f6211f56760ae99.png"},{"id":73201012,"identity":"3fa72d06-b180-47f7-a786-8118a0bc07ac","added_by":"auto","created_at":"2025-01-07 16:14:50","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":84672,"visible":true,"origin":"","legend":"\u003cp\u003ePredicting the accuracy of acetabular cup prosthesis.\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-5773489/v1/336ee4ff52eff78856c0ff7b.png"},{"id":73203166,"identity":"f9f441e6-4625-4593-8bd8-e6db9a0e9028","added_by":"auto","created_at":"2025-01-07 16:38:54","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2344702,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5773489/v1/ac0fe38b-3a08-4aed-9dd7-e09fbf4861f1.pdf"},{"id":73199873,"identity":"7aa92e69-4ff8-4985-ac71-26d637dd7f55","added_by":"auto","created_at":"2025-01-07 16:06:50","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":671134,"visible":true,"origin":"","legend":"","description":"","filename":"Appendix.docx","url":"https://assets-eu.researchsquare.com/files/rs-5773489/v1/30cf4aea9160d2f20f3bb7b0.docx"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eComparison of Artificial Intelligence and Traditional Methods in Preoperative Planning for Primary Total Hip Arthroplasty: A Systematic Review and Meta-Analysis.\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eTotal Hip Arthroplasty (THA) is a surgical procedure primarily used for elderly patients. With the increasing global average life expectancy and the growing proportion of elderly populations, the demand for THA continues to rise. In the United States, from 2020 to 2030, the number of primary and revision THA surgeries is expected to increase by approximately 50%\u003csup\u003e[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]\u003c/sup\u003e. In the United Kingdom, this number is projected to double by 2035\u003csup\u003e[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]\u003c/sup\u003e. Other Western countries, such as Canada, Denmark, and Australia, are also expected to see varying degrees of growth\u003csup\u003e[\u003cspan additionalcitationids=\"CR6\" citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]\u003c/sup\u003e. THA has existed in some form for over a century\u003csup\u003e[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]\u003c/sup\u003e, and in recent years, with technological advances and optimized techniques, THA has been dubbed the \"surgery of the 21st century,\" with its unique advantages being evident. THA has achieved significant success in alleviating joint pain, improving function, and enhancing quality of life\u003csup\u003e[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]\u003c/sup\u003e, with clinical outcomes being satisfactory, as the prosthetic survival rate over 15\u0026ndash;20 years can reach up to 90%\u003csup\u003e[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]\u003c/sup\u003e. As a result, THA has become the preferred treatment for end-stage degenerative joint diseases and hip trauma. By restoring normal anatomical relationships and the biomechanical stability of the hip joint, THA plays a crucial role in improving patients' quality of life\u003csup\u003e[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eHowever, THA is not without its challenges. Prosthetic joint infections, fractures, aseptic loosening, and dislocation\u003csup\u003e[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]\u003c/sup\u003e are the main causes of surgical failure\u003csup\u003e[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]\u003c/sup\u003e. These issues may not be entirely avoidable, even for experienced surgeons, and are particularly key obstacles in surgical planning for most general orthopedic surgeons. Therefore, accurate preoperative planning is crucial to improving surgical success rates and reducing complications.\u003c/p\u003e \u003cp\u003eTraditional preoperative planning mainly relies on two-dimensional templates\u003csup\u003e[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]\u003c/sup\u003e, which are limited by factors such as imaging angles and patient posture variations. As a result, the predicted outcomes are often influenced by the experience level of the surgeon. This reliance on experience often leads to considerable uncertainty, potentially resulting in surgical failure or poor short-term outcomes, which not only causes additional suffering for patients but also wastes valuable medical resources\u003csup\u003e[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]\u003c/sup\u003e. Hence, there is an urgent need for an efficient and accurate preoperative planning method.\u003c/p\u003e \u003cp\u003eThe rapid development of Artificial Intelligence (AI) has introduced new possibilities for preoperative planning in THA. AI's application in the medical field is continuously expanding, with breakthroughs in data analysis and image processing\u003csup\u003e[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]\u003c/sup\u003e. In the field of orthopedics, AI's application range is expanding, particularly in the research and treatment of hip diseases. Although current research on AI in THA preoperative planning is still in its early stages, with many studies being small-scale and lacking persuasive power, the potential of AI technology is becoming apparent\u003csup\u003e[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003e To better understand AI's role in THA preoperative planning, we performed a systematic review and meta-analysis of recent pertinent studies, focusing on the differences between AI and traditional methods in terms of preoperative planning accuracy and postoperative outcomes. By analyzing key indicators such as intraoperative blood loss, surgical time, postoperative leg length discrepancy (LLD), Harris scores, and prosthesis matching accuracy, we aim to address the following questions: Can AI significantly improve the accuracy of preoperative planning? Does it contribute to long-term patient recovery and a reduction in complications? What is its potential for future development? Our analysis may provide strong evidence to support clinical practice.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cp\u003e Our research follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. This study has been registered with the International Prospective Register of Systematic Reviews (PROSPERO) under registration number: CRD42024619714.\u003c/p\u003e \u003cp\u003eFor the experimental group in our study: The primary method of preoperative planning for total hip arthroplasty (THA) using artificial intelligence (AI) is an intelligent planning system based on 3D imaging and deep learning algorithms. This approach leverages AI technologies, such as deep learning algorithms and image processing techniques, to analyze patient imaging data (e.g., CT, MRI, or X-rays), enabling automated prosthesis selection, position planning, and surgical design. The preoperative planning models employed include AIHIP (Artificial Intelligence Preoperative Planning System for Total Hip Arthroplasty), 3D-AIHIP, robot-assisted THA, and CAD simulative (Computer-Aided Design Simulation).\u003c/p\u003e \u003cp\u003eFor the control group: The traditional method of preoperative planning for THA is primarily based on the template technique using two-dimensional X-ray images. This method involves obtaining standard X-rays of the patient's hip joint (e.g., anteroposterior pelvic view and lateral hip view), manually analyzing the anatomical features of the hip joint, and using a transparent prosthesis template superimposed on the X-ray to determine the appropriate prosthesis size and model. It also involves planning the implantation position, orientation, and angle of the prosthesis, as well as assessing limb length discrepancy to formulate a basic surgical plan.\u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eData Sources and Searches\u003c/h2\u003e \u003cp\u003eAccording to the established search criteria, we conducted a comprehensive search in databases such as Embase, Web of Science, PubMed, and the Cochrane Library. The search date ranged from the establishment of the database to October 30, 2024. We carefully verified the search methods used in each database to avoid any omissions. We also conducted a thorough summary of the search criteria and carefully reviewed the reference lists of relevant studies and reviews to find additional references. In these databases, we identified a total of 388 articles. After excluding 335 articles based on titles and abstracts, we further excluded 16 reviews, 19 duplicates, and 3 articles with inaccessible full texts or data that did not meet the research criteria, leaving 15 articles that were included in the qualitative analysis. The final selection included 2572 participants, with 1307 assigned to the experimental group and 1265 to the control group. The detailed search process can be found in the flowchart (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eInclusion and exclusion criteria\u003c/h3\u003e\n\u003cp\u003eThe method of using the Patient Intervention Compared Outcome of Interest (PICOS) model is as follows: In patients undergoing their first hip replacement surgery (P), artificial intelligence (AI) (I) is compared with traditional methods (C) to assess whether AI is more accurate in preoperative planning than traditional methods (O).\u003c/p\u003e \u003cp\u003eThe inclusion criteria for our study are: (1) The subjects are patients who require their first hip replacement surgery; (2) The experimental group uses AI models for preoperative planning; (3) The control group uses traditional methods for THA preoperative planning; (4) There are clear intraoperative and/or postoperative recorded indicators; (5) There are no statistically significant differences in baseline information such as age, BMI, preoperative Harris score, and preoperative leg length discrepancy across the studies.\u003c/p\u003e \u003cp\u003eExclusion criteria: (1) Conference abstracts, reviews, and articles that are registered but unpublished; (2) Articles for which the full text is inaccessible; (3) Articles with data that do not meet the research criteria; (4) Articles that are repetitively published (for repetitive publications, we selected the one with the most complete experimental data for statistical analysis).\u003c/p\u003e \u003cp\u003eAfter completing the search, all retrieved articles were first independently reviewed by one researcher (KW) for the titles and abstracts. To prevent selection bias due to factors such as personal fatigue or subjectivity, a second researcher (YD) verified the selection. Any disagreements encountered during this process were discussed, and if unresolved, a third mediator (GS) was involved to reach a consensus. Articles that were deemed appropriate for exclusion were eliminated. Then, the remaining articles were read in full, and those that met the inclusion criteria were finally included in our study.\u003c/p\u003e\n\u003ch3\u003eData Extraction\u003c/h3\u003e\n\u003cp\u003eIn the final included studies for statistical analysis, data extraction was conducted using Microsoft Excel to create data extraction tables. These tables were finalized after collaborative discussions among multiple team members. In addition to basic information such as title, year, primary author, age, BMI, sample size, and gender ratio, the extracted data primarily focused on intraoperative and postoperative details comparing the artificial intelligence group and the traditional method group. Key data points included the accuracy of acetabular and femoral matching predictions, operative time, intraoperative blood loss, postoperative leg length discrepancy, postoperative Harris scores, acetabular cup prediction accuracy, preoperative and postoperative VAS scores, postoperative inclination angle comparison, and length of hospital stay.\u003c/p\u003e \u003cp\u003eTwo researchers (KW and YD) independently conducted the data extraction process. Any discrepancies that occurred during the extraction process were addressed through discussion and a review by a third reviewer (YC). If disagreements persisted, more experienced researchers (DX) were consulted for guidance until a consensus was reached. (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.)\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eBasic information on preoperative planning for THA between the artificial intelligence group and the traditional methods group.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"13\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\"\u0026plusmn;\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\"\u0026plusmn;\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c11\" colnum=\"11\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c12\" colnum=\"12\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c13\" colnum=\"13\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eSource\u003c/p\u003e \u003cp\u003e(country)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSex\u003c/p\u003e \u003cp\u003e(male/\u003c/p\u003e \u003cp\u003efemale)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eAge: Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD\u003c/p\u003e \u003cp\u003e(year)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eFollow up(m)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c7\" namest=\"c6\"\u003e \u003cp\u003eBMI(kg/㎡)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c9\" namest=\"c8\"\u003e \u003cp\u003eInclusion model\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c11\" namest=\"c10\"\u003e \u003cp\u003ePostoperative Harris hip score\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c13\" namest=\"c12\"\u003e \u003cp\u003eNumber of participants\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003econtrol\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003econtrol\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eAI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003econtrol\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c10\"\u003e \u003cp\u003eAI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c11\"\u003e \u003cp\u003econtrol\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c12\"\u003e \u003cp\u003eAI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c13\"\u003e \u003cp\u003econtrol\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZheng, H.2024\u003c/p\u003e \u003cp\u003e(China)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e64.3\u0026thinsp;\u0026plusmn;\u0026thinsp;10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e \u003cp\u003e61.3\u0026thinsp;\u0026plusmn;\u0026thinsp;10.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e24.4\u0026thinsp;\u0026plusmn;\u0026thinsp;2.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e23.8\u0026thinsp;\u0026plusmn;\u0026thinsp;2.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eAIHIP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e2D model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e95.3\u0026thinsp;\u0026plusmn;\u0026thinsp;3.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e96.0\u0026thinsp;\u0026plusmn;\u0026thinsp;2.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c12\"\u003e \u003cp\u003e30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c13\"\u003e \u003cp\u003e30\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYang, W.2024\u003c/p\u003e \u003cp\u003e(China)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e193/247\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e56.6\u0026thinsp;\u0026plusmn;\u0026thinsp;14.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e \u003cp\u003e57\u0026thinsp;\u0026plusmn;\u0026thinsp;15.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eAIHIP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e2D model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e85.51\u0026thinsp;\u0026plusmn;\u0026thinsp;3.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e84.88\u0026thinsp;\u0026plusmn;\u0026thinsp;2.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c12\"\u003e \u003cp\u003e220\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c13\"\u003e \u003cp\u003e220\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eXie, H.2024\u003c/p\u003e \u003cp\u003e(China)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e16/87\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e42.3\u0026thinsp;\u0026plusmn;\u0026thinsp;12.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e \u003cp\u003e42.3\u0026thinsp;\u0026plusmn;\u0026thinsp;12.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e24.0\u0026thinsp;\u0026plusmn;\u0026thinsp;2.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e24.0\u0026thinsp;\u0026plusmn;\u0026thinsp;2.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eAIHIP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e2D model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c12\"\u003e \u003cp\u003e164\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c13\"\u003e \u003cp\u003e164\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSun, G. Y.2024\u003c/p\u003e \u003cp\u003e(China)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e32/40\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e62.2\u0026thinsp;\u0026plusmn;\u0026thinsp;10.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e \u003cp\u003e60.9\u0026thinsp;\u0026plusmn;\u0026thinsp;12.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e14.5\u0026thinsp;\u0026plusmn;\u0026thinsp;2.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eAIHIP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e2D model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e92.1\u0026thinsp;\u0026plusmn;\u0026thinsp;5.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e90.5\u0026thinsp;\u0026plusmn;\u0026thinsp;6.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c12\"\u003e \u003cp\u003e36\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c13\"\u003e \u003cp\u003e36\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMeipeng, M.2024\u003c/p\u003e \u003cp\u003e(China)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e28/32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e70.2\u0026thinsp;\u0026plusmn;\u0026thinsp;4.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e \u003cp\u003e69.8\u0026thinsp;\u0026plusmn;\u0026thinsp;5.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e5.5(4\u0026ndash;6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e22.8\u0026thinsp;\u0026plusmn;\u0026thinsp;4.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e23.1\u0026thinsp;\u0026plusmn;\u0026thinsp;3.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eAIHIP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e2D model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e81.43\u0026thinsp;\u0026plusmn;\u0026thinsp;4.65\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e77.51\u0026thinsp;\u0026plusmn;\u0026thinsp;3.28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c12\"\u003e \u003cp\u003e30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c13\"\u003e \u003cp\u003e30\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKai, Z.2024\u003c/p\u003e \u003cp\u003e(China)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e73/86\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e59.83\u0026thinsp;\u0026plusmn;\u0026thinsp;11.89\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e \u003cp\u003e61.90\u0026thinsp;\u0026plusmn;\u0026thinsp;13.88\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e13.72\u0026thinsp;\u0026plusmn;\u0026thinsp;5.65\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e22.20\u0026thinsp;\u0026plusmn;\u0026thinsp;1.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e22.50\u0026thinsp;\u0026plusmn;\u0026thinsp;1.75\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eAIHIP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e2D model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e92.74\u0026thinsp;\u0026plusmn;\u0026thinsp;3.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e91.81\u0026thinsp;\u0026plusmn;\u0026thinsp;3.52\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c12\"\u003e \u003cp\u003e80\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c13\"\u003e \u003cp\u003e79\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCheng, K.2024\u003c/p\u003e \u003cp\u003e(China)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e30/32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e60.5\u0026thinsp;\u0026plusmn;\u0026thinsp;13.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e \u003cp\u003e60.2\u0026thinsp;\u0026plusmn;\u0026thinsp;12.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eCAD simulative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003eroutine X-ray\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e81.9\u0026thinsp;\u0026plusmn;\u0026thinsp;6.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e74.7\u0026thinsp;\u0026plusmn;\u0026thinsp;11.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c12\"\u003e \u003cp\u003e31\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c13\"\u003e \u003cp\u003e31\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAnwar, A.2024\u003c/p\u003e \u003cp\u003e(China)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e52/65\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e62.3\u0026thinsp;\u0026plusmn;\u0026thinsp;10.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e \u003cp\u003e62.3\u0026thinsp;\u0026plusmn;\u0026thinsp;10.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e25.5\u0026thinsp;\u0026plusmn;\u0026thinsp;3.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e25.5\u0026thinsp;\u0026plusmn;\u0026thinsp;3.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eAIHIP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e2D model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c12\"\u003e \u003cp\u003e117\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c13\"\u003e \u003cp\u003e117\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZhang, S.2023\u003c/p\u003e \u003cp\u003e(China)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e19/48\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e46\u0026thinsp;\u0026plusmn;\u0026thinsp;18.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e \u003cp\u003e45\u0026thinsp;\u0026plusmn;\u0026thinsp;17.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e25.3\u0026thinsp;\u0026plusmn;\u0026thinsp;8.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e24.6\u0026thinsp;\u0026plusmn;\u0026thinsp;7.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003erobot-assisted THA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003emanual THA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e83\u0026thinsp;\u0026plusmn;\u0026thinsp;13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e83\u0026thinsp;\u0026plusmn;\u0026thinsp;11.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c12\"\u003e \u003cp\u003e36\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c13\"\u003e \u003cp\u003e31\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWu, L.2023\u003c/p\u003e \u003cp\u003e(China)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e101/60\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e57.6\u0026thinsp;\u0026plusmn;\u0026thinsp;10.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e \u003cp\u003e57.6\u0026thinsp;\u0026plusmn;\u0026thinsp;10.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e25.3\u0026thinsp;\u0026plusmn;\u0026thinsp;3.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e24.7\u0026thinsp;\u0026plusmn;\u0026thinsp;3.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e3D-AIHIP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e2D model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c12\"\u003e \u003cp\u003e95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c13\"\u003e \u003cp\u003e66\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWu, L.2023\u003c/p\u003e \u003cp\u003e(China)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e31/30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e58.2\u0026thinsp;\u0026plusmn;\u0026thinsp;9.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e \u003cp\u003e60.5\u0026thinsp;\u0026plusmn;\u0026thinsp;11.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026gt;12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e25.9\u0026thinsp;\u0026plusmn;\u0026thinsp;2.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e26.1\u0026thinsp;\u0026plusmn;\u0026thinsp;3.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e3D-AIHIP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e2D model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e86.38\u0026thinsp;\u0026plusmn;\u0026thinsp;4.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e84.37\u0026thinsp;\u0026plusmn;\u0026thinsp;5.30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c12\"\u003e \u003cp\u003e34\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c13\"\u003e \u003cp\u003e27\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMieradili, Maimaitiyiming\u003c/p\u003e \u003cp\u003e2023(China)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e10/18.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e59.3\u0026thinsp;\u0026plusmn;\u0026thinsp;16.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e \u003cp\u003e59.3\u0026thinsp;\u0026plusmn;\u0026thinsp;16.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eAIHIP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003eX-ray model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c12\"\u003e \u003cp\u003e28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c13\"\u003e \u003cp\u003e28\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eChen, X.2022\u003c/p\u003e \u003cp\u003e(China)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e61/59\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e47.62\u0026thinsp;\u0026plusmn;\u0026thinsp;15.30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e \u003cp\u003e53.75\u0026thinsp;\u0026plusmn;\u0026thinsp;16.10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e24.19\u0026thinsp;\u0026plusmn;\u0026thinsp;3.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e25.14\u0026thinsp;\u0026plusmn;\u0026thinsp;3.78\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eAIHIP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003eX-ray model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c12\"\u003e \u003cp\u003e60\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c13\"\u003e \u003cp\u003e60\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDing, X. Z.2021\u003c/p\u003e \u003cp\u003e(China)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e192/124\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e50.68\u0026thinsp;\u0026plusmn;\u0026thinsp;12.64\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e \u003cp\u003e50.68\u0026thinsp;\u0026plusmn;\u0026thinsp;12.64\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e25.07\u0026thinsp;\u0026plusmn;\u0026thinsp;3.20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e25.07\u0026thinsp;\u0026plusmn;\u0026thinsp;3.20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e3D-AIHIP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e2D model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c12\"\u003e \u003cp\u003e316\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c13\"\u003e \u003cp\u003e316\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWu, D.2020\u003c/p\u003e \u003cp\u003e(China)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e31/29\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e46.9\u0026thinsp;\u0026plusmn;\u0026thinsp;12.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e \u003cp\u003e47.0\u0026thinsp;\u0026plusmn;\u0026thinsp;21.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e3D-AIHIP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003efilm template\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c12\"\u003e \u003cp\u003e30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c13\"\u003e \u003cp\u003e30\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eSD: standard deviation; BMI: body mass index; NR: no reports; AIHIP: artificial intelligence preoperative planning system for total hip arthroplasty; CAD simulative: computer-aided design simulation; robot-assisted THA: robot-assisted total hip arthroplasty; 3D-AIHIP: artificial intelligence-assisted three-dimensional preoperative planning system for total hip arthroplasty.\u003c/p\u003e\n\u003ch3\u003eAssessment of Article Quality and Risk of Bias\u003c/h3\u003e\n\u003cp\u003eIn our study data, four studies were randomized controlled trials (RCTs), while the remaining studies were non-randomized controlled trials, including six retrospective cohort studies, two case-control studies, and three prospective cohort studies. Therefore, we adopted two different evaluation methods for assessing the quality of the articles. For the RCTs, the revised Cochrane risk-of-bias tool for randomized trials (ROB 2) was used, while the Newcastle-Ottawa Scale (NOS) was employed to evaluate the risk of bias for the non-randomized intervention studies included in our analysis.\u003c/p\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003eTo compare artificial intelligence (AI) with traditional methods in THA preoperative planning, we summarized the data for the experimental and control groups in each study and conducted a statistical analysis using a random-effects model. Since the collected data included both binary variables and continuous variables, in our analysis, we created forest plots that displayed either odds ratios (OR) for binary variables or mean differences (MD) for continuous variables. Heterogeneity was evaluated using the I\u0026sup2; statistic, where I\u0026sup2; \u0026gt; 50% was considered indicative of substantial heterogeneity, and I\u0026sup2; \u0026le; 50% indicated low heterogeneity.\u003c/p\u003e \u003cp\u003eWe illustrated the risk of bias using risk-of-bias plots and bar charts. The odds ratios (OR) or mean differences (MD) between the AI and traditional method groups were calculated with 95% confidence intervals (95% CI). The data were analyzed using RevMan version 5.4.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003eWe included 15 studies\u003csup\u003e[\u003cspan class=\"CitationRef\"\u003e21\u003c/span\u003e\u0026ndash;\u003cspan class=\"CitationRef\"\u003e35\u003c/span\u003e]\u003c/sup\u003e. Among the four randomized controlled trials (RCTs), two had no high-risk biases, while the other two each had one high-risk bias in terms of incomplete outcome data and selective reporting bias. Overall, the quality of the RCTs was quite high. For the 11 non-randomized controlled trials included, the NOS scores ranged from 4 to 8, with most scoring between 6 and 7. The NOS scale ranges from 0 to 9, and studies scoring above 4 are considered of moderate to high quality. Thus, the quality of the included studies was relatively high. Details can be found in the risk-of-bias assessment chart (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eSince the baseline characteristics during preoperative planning showed no statistical differences across studies, the postoperative statistical data are more comparable. We conducted multiple subgroup analyses of the intraoperative and postoperative data included in the studies. The results generally indicated that the AI group demonstrated advantages over the traditional method group in terms of preoperative planning accuracy, reducing surgery time, and lowering surgical risks.\u003c/p\u003e\n\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\n \u003ch2\u003ePrediction Accuracy for Acetabular and Femoral Sides:\u003c/h2\u003e\n \u003cp\u003eThe comparison is made between the preoperative planned prosthesis and the prosthesis actually used during the surgery. If the two are an exact match or differ by no more than one size (in THA, a difference of one size between the predicted acetabular or femoral prosthesis and the actual prosthesis used during the surgery is generally considered acceptable; the size difference of one size varies depending on the prosthesis type and the design specifications of different manufacturers, typically ranging from 1 to 2 millimeters), they are considered to have an accurate match.\u003c/p\u003e\n \u003cp\u003eThe predictive matching accuracy for the acetabular side\u003csup\u003e[22,23,25\u0026ndash;29,31\u0026minus;34]\u003c/sup\u003e in the AI group compared to the traditional method showed an OR of 0.26 (95% CI, 0.20\u0026ndash;0.34). For the femoral side prediction accuracy\u003csup\u003e[22,23,25\u0026ndash;29,31\u0026minus;34]\u003c/sup\u003e, the OR was 0.25 (95% CI, 0.19\u0026ndash;0.32). In both cases, the odds ratio (OR) for the experimental group compared to the control group was less than 1, indicating fewer prediction errors in the AI group, which suggests superior accuracy. (Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e.)\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\n \u003ch2\u003ePostoperative Leg Length Discrepancy (LLD):\u003c/h2\u003e\n \u003cp\u003eLLD refers to the length discrepancy between the lower limbs on both sides of the patient after THA. Specifically, it refers to the difference in skeletal length between the operated leg and the opposite leg, which is typically assessed through imaging examinations (such as X-rays or CT scans) or physical measurements (such as bedside measurement). A postoperative length difference of 1\u0026ndash;2 centimeters is generally considered acceptable. Nine studies\u003csup\u003e[\u003cspan class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e25\u003c/span\u003e\u0026ndash;\u003cspan class=\"CitationRef\"\u003e28\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e30\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e33\u003c/span\u003e]\u003c/sup\u003e reported postoperative leg length discrepancy (LLD). The mean difference (MD) between the experimental group and the control group was \u0026minus;\u0026thinsp;0.49 (95% CI, -0.59 to -0.39). This indicates that preoperative planning with artificial intelligence resulted in lower leg length discrepancies postoperatively compared to the control group. The surgical error margin in the experimental group was smaller, leading to better postoperative adaptation for patients. (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e.)\u003c/p\u003e\n \u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\n \u003ch2\u003eSurgical Time and Intraoperative Blood Loss:\u003c/h2\u003e\n \u003cp\u003eEight studies included data on surgical time\u003csup\u003e[\u003cspan class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e29\u003c/span\u003e\u0026ndash;\u003cspan class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e33\u003c/span\u003e]\u003c/sup\u003e, and five studies\u003csup\u003e[\u003cspan class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e33\u003c/span\u003e]\u003c/sup\u003e reported intraoperative blood loss. The mean differences (MD) between the experimental group and the control group were \u0026minus;\u0026thinsp;16.07 (95% CI, -18.00 to -14.00) and \u0026minus;\u0026thinsp;45.91 (95% CI, -61.03 to -30.78), respectively. These results indicate that preoperative planning with artificial intelligence (AI) for THA, compared to traditional methods, resulted in shorter surgical times and reduced intraoperative blood loss.\u003c/p\u003e\n \u003cp\u003eShorter surgical times can effectively lower surgical risks, including reducing the likelihood of infections caused by prolonged exposure. Decreased intraoperative blood loss reduces patient life risk and avoids unnecessary blood transfusions, thereby mitigating transfusion-related risks such as antibody reactions. Furthermore, this reduction helps alleviate blood supply shortages, ensuring that more patients in critical need of transfusions can access the necessary resources. (Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e.)\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\n \u003ch2\u003ePostoperative Harris Score (HHS):\u003c/h2\u003e\n \u003cp\u003eHHS is a standardized scoring system used to assess the recovery of hip joint function in patients after total hip arthroplasty (THA). This scoring system takes into account the patient\u0026apos;s level of pain, activity ability, gait, joint function and range, as well as the ability to perform daily activities related to the hip joint. The total score ranges from 0 to 100 and is interpreted as follows:\u003c/p\u003e\n \u003cul\u003e\n \u003cli\u003e\n \u003cp\u003e90\u0026ndash;100 points: indicates good function with almost no issues.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003e80\u0026ndash;89 points: indicates fair function, with possible mild symptoms.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003e70\u0026ndash;79 points: indicates moderate function, with some discomfort and limitations.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eBelow 60 points: indicates poor function, with significant pain or functional impairment.\u003c/p\u003e\n \u003c/li\u003e\n \u003c/ul\u003e\n \u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n \u003cp\u003eIn other words, a higher score indicates better postoperative recovery.\u003c/p\u003e\n \u003cp\u003eIn this research, eight studies\u003csup\u003e[\u003cspan class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e27\u003c/span\u003e\u0026ndash;\u003cspan class=\"CitationRef\"\u003e31\u003c/span\u003e]\u003c/sup\u003e included the postoperative Harris scores (and the preoperative Harris scores of each study showed no statistical difference, indicating that the postoperative Harris scores are more comparable). The mean difference (MD) of postoperative Harris scores in these 8 studies was 0.83 (95% CI, 0.38\u0026ndash;1.28), which means that the average postoperative Harris score in the AI group was higher than in the traditional method group. The I\u0026sup2; was 70%, indicating significant heterogeneity, with a P-value of \u0026lt;\u0026thinsp;0.05, which shows a statistically significant difference. This suggests that preoperative planning using AI results in less pain and better functional recovery postoperatively, leading to a more comfortable treatment experience for the patients. (Fig. \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003e.)\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\n \u003ch2\u003eAccuracy of Acetabular Cup Prosthesis Prediction:\u003c/h2\u003e\n \u003cp\u003eThe accuracy of acetabular cup prosthesis prediction refers to the degree of agreement between the size, position, and angle of the acetabular cup prosthesis predicted in the preoperative plan and the actual prosthesis implanted during the surgery in THA. If the size, position, and angle deviations of the acetabular cup prosthesis are within an acceptable range (which may vary depending on the design specifications of different manufacturers, with a general range being: size deviation of 1-2mm, position deviation of 5mm, and angle deviation of 5\u0026deg;), the preoperative planned prosthesis is considered to match the actual implanted prosthesis accurately. The higher the accuracy, the smaller the deviation between the preoperative plan and the actual intraoperative operation, resulting in better surgical outcomes.\u003c/p\u003e\n \u003cp\u003eThere are three studies\u003csup\u003e[\u003cspan class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e28\u003c/span\u003e]\u003c/sup\u003e evaluated the accuracy of acetabular cup prosthesis prediction, with an odds ratio (OR) of 0.82 (95% CI, 0.51\u0026ndash;1.34) and P\u0026thinsp;=\u0026thinsp;0.43. Although the small sample size limits the persuasiveness of the findings, the overall trend suggests that the AI group demonstrates superior accuracy in predicting acetabular prosthesis outcomes compared to the traditional method group. (Fig. \u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003e.)\u003c/p\u003e\n \u003cp\u003eOverall, the AI group has indeed brought significant surprises in the field of healthcare. However, AI still cannot guarantee complete accuracy, but its potential for development is vast.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"Discussion","content":"\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eAdvantages and challenges of preoperative planning assisted by AI in THA:\u003c/h2\u003e \u003cp\u003eTotal Hip Arthroplasty (THA) is currently the primary surgical method for treating persistent pain and functional limitations caused by advanced hip joint diseases. With the increasing global average lifespan and continuous advancements in surgical technology, the age range of patients undergoing THA has expanded significantly, showing a growing trend of polarization\u003csup\u003e[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]\u003c/sup\u003e. This shift not only imposes higher demands on surgical complexity but also profoundly impacts traditional medical decision-making models\u003csup\u003e[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]\u003c/sup\u003e. It is projected that the number of patients requiring THA annually will increase substantially in the future.\u003c/p\u003e \u003cp\u003eAlthough THA is one of the most successful procedures in orthopedics, its outcomes can still be significantly affected by the choice of prosthesis model and the precision of implantation. Improper prosthesis matching can lead to intraoperative and postoperative complications such as early prosthesis loosening, dislocation, or even surgical failure, ultimately affecting the longevity of the prosthesis and the patient's mobility\u003csup\u003e[\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]\u003c/sup\u003e. Precise matching between the prosthesis and bone, optimal implant positioning, and superior friction interfaces are key to constructing a stable mechanical structure and prolonging the lifespan of the prosthesis. To achieve these goals, comprehensive and accurate preoperative planning is crucial. Preoperative planning optimizes surgical processes, reduces unnecessary steps during surgery, shortens operation time, lowers risks, and promotes postoperative recovery\u003csup\u003e[\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]\u003c/sup\u003e. In recent years, with the rise of digital orthopedic technology, three-dimensional preoperative planning based on CT or X-ray imaging data has demonstrated significant advantages in the treatment of hip joint diseases. Research has demonstrated that AI-assisted 3D preoperative design techniques can facilitate more accurate, safer, and consistently reproducible acetabular prosthesis implantation \u003csup\u003e[\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cdiv id=\"Sec16\" class=\"Section3\"\u003e \u003ch2\u003eLimitations of traditional preoperative planning and the rise of artificial intelligence:\u003c/h2\u003e \u003cp\u003eTraditional manual imaging review heavily relies on the personal experience of physicians, leading to significant variations in preoperative planning accuracy. In contrast, Artificial Intelligence (AI) with deep learning technology can integrate diagnostic expertise from multiple specialists, offering more objective and precise diagnostic and planning recommendations through big data analysis. Studies have shown that, compared to traditional methods, AI excels in the predictive accuracy of prosthesis models and implant angles. For instance, AI-assisted three-dimensional preoperative planning achieves prediction accuracies of 90% for acetabular cups and 83% for femoral stems, significantly higher than the 57% and 53% achieved by traditional two-dimensional X-ray planning\u003csup\u003e[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]\u003c/sup\u003e.In our study, the AI-assisted group not only significantly improved prosthesis matching accuracy but also outperformed traditional methods in several key metrics, including shorter surgery time, reduced intraoperative blood loss, smaller postoperative leg length discrepancy (LLD), and higher postoperative Harris scores. Particularly in prosthesis implant angle planning, AI, combined with 3D printing technology and personalized navigation templates, can reconstruct individualized surgical models, optimizing implant paths and angles to minimize fluoroscopy use and surgical trauma\u003csup\u003e[\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eThe potential of AI in managing THA postoperative complications:\u003c/h2\u003e \u003cp\u003eThe hip joint bears the majority of daily weight-bearing activities, and improper prosthesis matching can lead to postoperative complications such as early dislocation, pain, and loosening\u003csup\u003e[\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]\u003c/sup\u003e. AI-assisted preoperative planning enables more accurate predictions of prosthesis models and implant angles, achieving optimal matching between the prosthesis and bone, thereby reducing the incidence of dislocations and other complications, while enhancing joint stability and prosthesis longevity\u003csup\u003e[\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eLLD is one of the most common postoperative complications, considered a primary cause of postoperative pain, gait instability, and aseptic loosening. In severe cases, it may necessitate revision surgery within a short period\u003csup\u003e[\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]\u003c/sup\u003e. Statistical analysis indicates that patients with AI-assisted planning exhibit significantly lower postoperative LLD means compared to the traditional method group (MD = -0.49, 95% CI: -0.59 to -0.39, P\u0026thinsp;\u0026lt;\u0026thinsp;0.05), demonstrating statistically significant differences.\u003c/p\u003e \u003cp\u003eFurthermore, regarding the precision of acetabular prosthesis implantation, the proportion of AI-planned acetabular prostheses positioned within Lewinnek and Callanan safe zones reached 86.32% and 83.2%, respectively, significantly exceeding the 72.73% and 69.7% achieved by traditional methods. This highlights AI's advantages in planning complex cases.\u003c/p\u003e \u003cdiv id=\"Sec18\" class=\"Section3\"\u003e \u003ch2\u003eChallenges and future perspectives:\u003c/h2\u003e \u003cp\u003eAlthough AI demonstrates numerous advantages in preoperative planning for Total Hip Arthroplasty (THA), certain challenges remain in its clinical application. First, current studies are limited in sample size and follow-up duration, and the lack of long-term follow-up data makes it difficult to comprehensively evaluate AI's impact on prosthesis longevity and functional recovery. Second, AI's development in the medical field is still in its early stages, and its reliability and generalizability in complex cases require further validation.\u003c/p\u003e \u003cp\u003eIn conclusion, AI offers more precise and efficient solutions for THA preoperative planning, with remarkable potential to optimize surgical workflows, enhance postoperative functional recovery, and reduce complications. However, the widespread adoption of AI will require large-scale, long-term follow-up studies and multidisciplinary collaboration to continually refine the technology and optimize its application strategies, paving the way for AI to achieve greater maturity in the field of orthopedics.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section3\"\u003e \u003ch2\u003eLimitations\u003c/h2\u003e \u003cp\u003eOur study has several limitations that warrant attention and improvement in future research. First, the follow-up duration in the included studies was limited, with the longest recorded follow-up being 19.37 months\u003csup\u003e[\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]\u003c/sup\u003e and the shortest only 3 months. Although the AI-assisted preoperative planning group demonstrated superior performance in prosthesis placement accuracy and correction of postoperative leg length discrepancy (LLD), it remains unclear whether these advantages in prosthesis accuracy can be sustained over medium- to long-term follow-ups. Therefore, future research should involve larger sample sizes and longer follow-up periods to validate the long-term benefits of these advantages. Second, all the studies we included were conducted in China. Although this reflects the objective findings of our screening process, the single-region research background may pose certain limitations. To ensure the broader applicability of our findings, we adhered to strict bias risk assessment methods, using the Cochrane risk of bias tool for randomized controlled trials (RCTs) and the Newcastle-Ottawa Scale (NOS) for non-randomized controlled trials (non-RCTs). Fortunately, the overall quality of the included studies was relatively high. However, research conducted in a single region may not fully represent the situation in different parts of the world. Therefore, we look forward to the inclusion of more relevant studies from diverse regions in the future to enrich the research content and enhance the reliability and generalizability of the conclusions.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Conclusion","content":"\u003cp\u003eCompared to traditional preoperative planning methods, artificial intelligence (AI)-assisted preoperative planning demonstrates significant advantages. Firstly, AI provides a more intuitive visualization of the anatomical structure of the affected area, combined with three-dimensional technology for preoperative simulations. This enhances surgical precision and reduces procedural complexity. Additionally, the application of AI decreases the need for intraoperative fluoroscopy and repeated prosthesis measurements. These benefits not only shorten the operation time and minimize intraoperative blood loss but also effectively lower surgical risks and trauma, promoting faster postoperative recovery, reducing the incidence of complications, and ultimately preventing the wastage of medical resources.\u003c/p\u003e \u003cp\u003eIn terms of prosthesis parameter prediction and matching accuracy, AI surpasses traditional methods. This benefit aids in ensuring both the initial stability and the long-term durability of the prosthesis \u003csup\u003e[\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]\u003c/sup\u003e. However, to further confirm the reliability and feasibility of these conclusions, future studies need larger-scale randomized controlled trials (RCTs) and extended follow-up periods to provide stronger evidence.AI has demonstrated extensive potential in the medical field, with vast opportunities for further development in both depth and breadth. However, it is crucial to recognize that technology is a double-edged sword. No matter how advanced AI becomes, its development and application should always be guided by principles of humanistic ethics, ensuring that it serves the health and well-being of humanity and society.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eSupplementary Information\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSupplementary materials mentioned in the article can be found in the appendix.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026rsquo; Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eKaiyong Wang, Yupei Dai and Di Xue wrote the first draft of the article, Kaiyong Wang,Yupei Dai and Guohang Shen made revisions to the manuscript, as well as language polishing of the article, Guohang Shen, Yang Chen assisted in the process of data extraction and entry, Kaiyong Wang,Yupei Dai and Yang Chen provided reference materials, and Kaiyong Wang and Di Xue finally read and confirmed the manuscript of the article.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNatural Science Foundation of Ningxia Hui Autonomous Region（2024AAC03601，2024AAC03663，2024AAC03665），Ningxia Medical University project（XT2023035），The central government of Ningxia Hui Autonomous Region guides the special project of local science and technology development（2024FRD05048，2024FRD05108），Key R\u0026amp;D project of Ningxia Hui Autonomous Region（2022BEG03126，2022BEG03169）.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthical Statement:\u003cbr\u003e\u003c/strong\u003eThis study falls within the scope of retrospective research and does not involve direct human participants, animal experiments, or the collection and use of sensitive data. Therefore, no ethical issues are present.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNone.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of Interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that there are no conflicts of interest associated with the research presented in this paper.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eLan Z, Lin X, Xue D et al (2024) Can Bisphosphonate Therapy Reduce Overall Mortality in Patients With Osteoporosis? A Meta-analysis of Randomized Controlled Trials. Clinical orthopaedics and related research\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.1097/corr.0000000000003204\u003c/span\u003e\u003cspan address=\"10.1097/corr.0000000000003204\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKurtz S, Ong K, Lau E et al (2007) Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am Vol 89(4):780\u0026ndash;785. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.2106/jbjs.F.00222\u003c/span\u003e\u003cspan address=\"10.2106/jbjs.F.00222\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSloan M, Premkumar A, Sheth NP (2018) Projected Volume of Primary Total Joint Arthroplasty in the U.S., 2014 to 2030. J Bone Joint Surg Am Vol 100:17, 1455\u0026ndash;1460. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.2106/jbjs.17.01617\u003c/span\u003e\u003cspan address=\"10.2106/jbjs.17.01617\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCulliford D, Maskell J, Judge A et al (2015) Future projections of total hip and knee arthroplasty in the UK: results from the UK Clinical Practice Research Datalink. Osteoarthr Cartil 23:4, 594\u0026ndash;600. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.1016/j.joca.2014.12.022\u003c/span\u003e\u003cspan address=\"10.1016/j.joca.2014.12.022\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSharif B, Kopec J, Bansback N et al (2015) Projecting the direct cost burden of osteoarthritis in Canada using a microsimulation model. Osteoarthr Cartil 23:10, 1654\u0026ndash;1663. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.1016/j.joca.2015.05.029\u003c/span\u003e\u003cspan address=\"10.1016/j.joca.2015.05.029\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNemes S, Gordon M, Rogmark C et al (2014) Projections of total hip replacement in Sweden from 2013 to 2030. Acta Orthop 85:3, 238\u0026ndash;243. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.3109/17453674.2014.913224\u003c/span\u003e\u003cspan address=\"10.3109/17453674.2014.913224\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAckerman IN, Bohensky MA, Zomer E et al (2019) The projected burden of primary total knee and hip replacement for osteoarthritis in Australia to the year 2030. BMC Musculoskelet Disord 20(1):90. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.1186/s12891-019-2411-9\u003c/span\u003e\u003cspan address=\"10.1186/s12891-019-2411-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKnight SR, Aujla R, Biswas SP (2011) Total Hip Arthroplasty - over 100 years of operative history. Orthop Rev 3:e16. 2\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.4081/or.2011.e16\u003c/span\u003e\u003cspan address=\"10.4081/or.2011.e16\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLearmonth ID, Young C, Rorabeck C (2007) The operation of the century: total hip replacement. Lancet (London England) 370:9597, 1508\u0026ndash;1519. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.1016/s0140-6736(07)60457-7\u003c/span\u003e\u003cspan address=\"10.1016/s0140-6736(07)60457-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eImamura M, Munro NA, Zhu S et al (2012) Single mini-incision total hip replacement for the management of arthritic disease of the hip: a systematic review and meta-analysis of randomized controlled trials. J Bone Joint Surg Am Vol 94:20, 1897\u0026ndash;1905. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.2106/jbjs.K.00495\u003c/span\u003e\u003cspan address=\"10.2106/jbjs.K.00495\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlp NB, Akdağ G, Erdoğan F (2020) Long-term results of total hip arthroplasty in developmental dysplasia of hip patients. Joint Dis Relat Surg 31:2, 298\u0026ndash;305. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.5606/ehc.2020.74412\u003c/span\u003e\u003cspan address=\"10.5606/ehc.2020.74412\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEvans JT, Evans JP, Walker RW et al (2019) How long does a hip replacement last? A systematic review and meta-analysis of case series and national registry reports with more than 15 years of follow-up. Lancet (London England) 393:10172, 647\u0026ndash;654. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.1016/s0140-6736(18)31665-9\u003c/span\u003e\u003cspan address=\"10.1016/s0140-6736(18)31665-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKumar Vyas A, Gupta A, Patni P et al (2020) Evaluation of results of various tendon transfers in high and low radial nerve palsy. J Clin Orthop trauma 11:4, 614\u0026ndash;619. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.1016/j.jcot.2020.05.023\u003c/span\u003e\u003cspan address=\"10.1016/j.jcot.2020.05.023\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKarachalios T, Komnos G, Koutalos A (2018) Total hip arthroplasty: Survival and modes of failure. EFORT open reviews 3:5, 232\u0026ndash;239. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.1302/2058-5241.3.170068\u003c/span\u003e\u003cspan address=\"10.1302/2058-5241.3.170068\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDi Martino A, Brunello M, Bordini B et al (2023) Unstable Total Hip Arthroplasty: Should It Be Revised Using Dual Mobility Implants? A Retrospective Analysis from the R.I.P.O. Registry. J Clin Med 12:2. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.3390/jcm12020440\u003c/span\u003e\u003cspan address=\"10.3390/jcm12020440\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eViceconti M, Ascani D, Mazz\u0026agrave; C (2019) Pre-operative prediction of soft tissue balancing in knee arthoplasty part 1: Effect of surgical parameters during level walking. J Orthop research: official publication Orthop Res Soc 37:7, 1537\u0026ndash;1545. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.1002/jor.24289\u003c/span\u003e\u003cspan address=\"10.1002/jor.24289\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHo CWG, Tan BK, Ong BH et al (2021) Customized Two-Dimensional Computed Tomography-Guided Preoperative Pectus Bar Shaping. Annals of plastic surgery 87. 6676\u0026ndash;680. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.1097/sap.0000000000002892\u003c/span\u003e\u003cspan address=\"10.1097/sap.0000000000002892\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGu LC, Peng Y, Zhang Y et al (2024) Enhancing treatment outcomes for Acute Periprosthetic Hip Joint infection: optimizing debridement, antibiotics, and Implant Retention through vacuum sealing drainage in the deep tissue. Arch Orthop Trauma Surg 145:1, 54. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.1007/s00402-024-05649-z\u003c/span\u003e\u003cspan address=\"10.1007/s00402-024-05649-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShen D, Wu G, Suk HI (2017) Deep Learning in Medical Image Analysis. Annu Rev Biomed Eng 19:221\u0026ndash;248. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.1146/annurev-bioeng-071516-044442\u003c/span\u003e\u003cspan address=\"10.1146/annurev-bioeng-071516-044442\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZaharchuk G (2019) Next generation research applications for hybrid PET/MR and PET/CT imaging using deep learning. Eur J Nucl Med Mol Imaging 46:13, 2700\u0026ndash;2707. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.1007/s00259-019-04374-9\u003c/span\u003e\u003cspan address=\"10.1007/s00259-019-04374-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZheng H, Feng E, Xiao Y et al (2024) Is AI 3D-printed PSI an accurate option for patients with developmental dysplasia of the hip undergoing THA? BMC Musculoskelet Disord 25(1):308. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.1186/s12891-024-07449-3\u003c/span\u003e\u003cspan address=\"10.1186/s12891-024-07449-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAnwar A, Zhang Y, Zhang Z et al (2024) Artificial intelligence technology improves the accuracy of preoperative planning in primary total hip arthroplasty. Asian J Surg 47:7, 2999\u0026ndash;3006. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.1016/j.asjsur.2024.01.133\u003c/span\u003e\u003cspan address=\"10.1016/j.asjsur.2024.01.133\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang W, Gao T, Liu X et al (2024) Clinical application of artificial intelligence-assisted three-dimensional planning in direct anterior approach hip arthroplasty. Int Orthop 48:3, 773\u0026ndash;783. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.1007/s00264-023-06029-9\u003c/span\u003e\u003cspan address=\"10.1007/s00264-023-06029-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXie H, Yi J, Huang Y et al (2024) Application and evaluation of artificial intelligence 3D preoperative planning software in developmental dysplasia of the hip. J Orthop Surg Res 19:1, 176. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.1186/s13018-024-04588-0\u003c/span\u003e\u003cspan address=\"10.1186/s13018-024-04588-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMeipeng M, Jin W, Rafi U et al (2024) Role and significance of artificial intelligence preoperative planning in total hip arthroplasty. Chin J Tissue Eng Res 28:9, 1372\u0026ndash;1377. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.12307/2023.798\u003c/span\u003e\u003cspan address=\"10.12307/2023.798\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu L, Zhao X, Lu ZD et al (2023) Accuracy analysis of artificial intelligence-assisted three-dimensional preoperative planning in total hip replacement. Joint Dis Relat Surg 34:3, 537\u0026ndash;547. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.52312/jdrs.2023.1059\u003c/span\u003e\u003cspan address=\"10.52312/jdrs.2023.1059\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSun GY, Jiang YK, Li T et al (2024) AI-HIP system for prosthesis size, global femoral offset and osteotomy in total hip arthroplasty. Zhongguo gu shang = China J Orthop Traumatol 37:9, 848\u0026ndash;854. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.12200/j.issn.1003-0034.20240276\u003c/span\u003e\u003cspan address=\"10.12200/j.issn.1003-0034.20240276\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKai Z, Zhuotao G, Qiaoqiao M et al (2024) Accuracy and influencing factor of artificial intelligence planning system in patients undergoing total hip arthroplasty. Chin J Tissue Eng Res 28:12, 1863\u0026ndash;1868. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.12307/2024.026\u003c/span\u003e\u003cspan address=\"10.12307/2024.026\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCheng K, Zhu HT, Peng YH et al (2024) To further incorporate computer-aided designs to improve preoperative planning in total hip arthroplasty: a cohort study. Front Surg 11. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.3389/fsurg.2024.1345261\u003c/span\u003e\u003cspan address=\"10.3389/fsurg.2024.1345261\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang S, Liu Y, Ma M et al (2023) Is Robotic-Assisted Technology Still Accurate in Total Hip Arthroplasty for Fibrous-Fused Hips? J Arthroplast 38:1, 129\u0026ndash;134. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.1016/j.arth.2022.07.023\u003c/span\u003e\u003cspan address=\"10.1016/j.arth.2022.07.023\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu L, Yang XC, Wu J et al (2023) Short-term outcome of artificial intelligence-assisted preoperative three-dimensional planning of total hip arthroplasty for developmental dysplasia of the hip compared to traditional surgery. 34:571\u0026ndash;582. Joint diseases and related surgery\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.52312/jdrs.2023.1076\u003c/span\u003e\u003cspan address=\"10.52312/jdrs.2023.1076\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMieradili M, Yilihamujiang W, Sun R et al (2023) : Application of artificial intelligence preoperative planning system in total hip arthroplasty for adult developmental dysplasia of the hip. Zhongguo xiu fu chong jian wai ke za zhi\u0026thinsp;=\u0026thinsp;Zhongguo xiufu chongjian waike zazhi\u0026thinsp;=\u0026thinsp;Chinese journal of reparative and reconstructive surgery 37, 1, 25\u0026ndash;30. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.7507/1002-1892.202209098\u003c/span\u003e\u003cspan address=\"10.7507/1002-1892.202209098\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen X, Liu X, Wang Y et al (2022) Development and Validation of an Artificial Intelligence Preoperative Planning System for Total Hip Arthroplasty. Front Med 9:841202. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.3389/fmed.2022.841202\u003c/span\u003e\u003cspan address=\"10.3389/fmed.2022.841202\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDing XZ, Zhang BS, Li WA et al (2021) Value of preoperative three-dimensional planning software (AI-HIP) in primary total hip arthroplasty: a retrospective study. J Int Med Res 49:11. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.1177/03000605211058874\u003c/span\u003e\u003cspan address=\"10.1177/03000605211058874\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu D, Liu X, Zhang Y et al (2020) : Research and application of artificial intelligence based three-dimensional preoperative planning system for total hip arthroplasty. Zhongguo xiu fu chong jian wai ke za zhi\u0026thinsp;=\u0026thinsp;Zhongguo xiufu chongjian waike zazhi\u0026thinsp;=\u0026thinsp;Chinese journal of reparative and reconstructive surgery 34, 9, 1077\u0026ndash;1084. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.7507/1002-1892.202005007\u003c/span\u003e\u003cspan address=\"10.7507/1002-1892.202005007\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCallanan MC, Jarrett B, Bragdon CR et al (2011) The John Charnley Award: risk factors for cup malpositioning: quality improvement through a joint registry at a tertiary hospital. Clin Orthop Relat Res 469:2, 319\u0026ndash;329. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.1007/s11999-010-1487-1\u003c/span\u003e\u003cspan address=\"10.1007/s11999-010-1487-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang J, Min L, Lu M et al (2020) : What are the Complications of Three-dimensionally Printed, Custom-made, Integrative Hemipelvic Endoprostheses in Patients with Primary Malignancies Involving the Acetabulum, and What is the Function of These Patients? Clinical orthopaedics and related research 478, 11, 2487\u0026ndash;2501. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.1097/corr.0000000000001297\u003c/span\u003e\u003cspan address=\"10.1097/corr.0000000000001297\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOfa SA, Ross AJ, Ross BJ et al (2021) : Complication Rates of Hemiarthroplasty Conversion to Total Hip Arthroplasty Versus Primary Total Hip Arthroplasty. Orthopedic reviews 13, 2, 25539. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.52965/001c.25539\u003c/span\u003e\u003cspan address=\"10.52965/001c.25539\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Natural Science Foundation of Ningxia Hui Autonomous Region（2024AAC03601，2024AAC03663，2024AAC03665），","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Artificial intelligence (AI), Traditional methods, Total hip arthroplasty/Total hip replacement, Preoperative planning, First-time.","lastPublishedDoi":"10.21203/rs.3.rs-5773489/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5773489/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe application of artificial intelligence (AI) in orthopedics is becoming increasingly widespread, particularly in the diagnosis and treatment of hip-related diseases. Although AI-assisted total hip arthroplasty (THA) techniques have reached a relatively mature stage, their specific role in preoperative planning for THA remains in the research phase. Current studies are generally small in scale, and their findings appear somewhat fragmented, making it difficult to draw definitive conclusions. Against this backdrop, a systematic review and meta-analysis on the application of AI in THA preoperative planning may provide a more comprehensive and rational answer.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eQuestions/purposes\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCompared to traditional methods, does artificial intelligence (AI) offer more and better advantages in preoperative planning for patients undergoing primary total hip arthroplasty (THA)? Does it possess potential for future development?\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe conducted a comprehensive and systematic search in the PubMed, Embase, Web of Science, and Cochrane Library databases, covering the period from their inception to October 30, 2024. This study adheres to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and has been registered in PROSPERO\u003csup\u003e[1]\u003c/sup\u003e. The included studies focused on patients undergoing primary total hip arthroplasty (THA), with the experimental group using artificial intelligence (AI) for preoperative planning and the control group employing traditional planning methods. We excluded the following: papers published on preprint servers, unpublished studies, conference abstracts, and studies registered on ClinicalTrials.gov but not yet published. Ultimately, data were extracted from 15 eligible studies.\u003c/p\u003e\n\u003cp\u003eTo assess the methodological quality of the studies, we applied bias risk assessment methods based on the type of study. The revised Cochrane Risk of Bias tool was employed to assess potential bias in randomized controlled trials (RCTs). For non-randomized controlled trials, including retrospective cohort studies, retrospective case-control studies, and prospective cohort studies, we employed the Newcastle-Ottawa Scale (NOS) for bias risk assessment. Due to the high heterogeneity among studies (I² \u0026gt; 50%), a random-effects model was used for the analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn the 15 studies that met the inclusion criteria, a total of 2572 participants were included. These patients required primary total hip arthroplasty (THA) due to various hip diseases. Among them, 1307 patients in the experimental group used artificial intelligence (AI) for preoperative planning, while 1265 patients in the control group used traditional methods. There were no statistically significant differences in the baseline characteristics of the included patients (such as age, BMI, preoperative leg length discrepancy, and preoperative Harris score) (P≥0.05), which ensures the reliability of the predictive results.\u003c/p\u003e\n\u003cp\u003eAccording to the data summary and analysis, compared with traditional methods, AI showed superior performance in the following aspects: the odds ratio (OR) for acetabular component matching accuracy was 0.26 (95% CI, 0.20–0.34; P=0.009; I²=58%), and for femoral component matching accuracy, the OR was 0.25 (95% CI, 0.19–0.32; P=0.66; I²=0%). The matching accuracy was defined with a size difference as the acceptable margin of error. The mean difference (MD) for postoperative leg length discrepancy was -0.49 (95% CI, -0.59 to -0.39; P\u0026lt;0.0001; I²=77%), the MD for surgical time was -16.07 (95% CI, -18.00 to -14.14; P\u0026lt;0.00001; I²=96%), the MD for intraoperative blood loss was -45.91 (95% CI, -61.03 to -30.78; P=0.04; I²=61%), and the MD for postoperative Harris score was 0.83 (95% CI, 0.38–1.28; P=0.001; I²=70%). In addition, the OR for acetabular cup prosthesis prediction accuracy was 0.82 (95% CI, 0.51–1.34; P=0.0001; I²=89%), and the overall average prediction accuracy had an OR of 0.25 (95% CI, 0.18–0.35; P=0.93; I²=0%).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe results of this systematic review and meta-analysis indicate that artificial intelligence (AI) performs comparably to, or even better than, traditional methods in preoperative planning for hip arthroplasty. Compared with traditional methods, the AI group demonstrated advantages such as reducing surgical time, minimizing intraoperative blood loss, lowering surgical risks, and decreasing surgical trauma. These benefits help promote rapid postoperative recovery, shorten hospital stays, and reduce the occurrence of complications. Additionally, patients in the AI group had higher postoperative Harris scores, less postoperative pain, faster functional recovery, and better postoperative adaptation. AI-assisted preoperative planning for total hip arthroplasty (THA) also improves the accuracy of hip component matching prediction, reduces the likelihood of errors in clinical decision-making, effectively alleviates tensions in the doctor-patient relationship, and reduces the waste of medical resources.\u003c/p\u003e","manuscriptTitle":"Comparison of Artificial Intelligence and Traditional Methods in Preoperative Planning for Primary Total Hip Arthroplasty: A Systematic Review and Meta-Analysis.","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-01-07 16:06:45","doi":"10.21203/rs.3.rs-5773489/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"707f1a80-685e-41bc-a01e-3f44e1f5d785","owner":[],"postedDate":"January 7th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":42425216,"name":"Surgery"},{"id":42425217,"name":"Orthopedics"}],"tags":[],"updatedAt":"2025-01-07T16:06:45+00:00","versionOfRecord":[],"versionCreatedAt":"2025-01-07 16:06:45","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5773489","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5773489","identity":"rs-5773489","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}